The meeting carries particular significance amid recent policy disagreements and political tensions between the two over AI safety, technological expansion and government regulation.

AI Regulation and the Pentagon Dispute

AI safety and policy are expected to be the main topics of discussion. To mitigate risks in the technology sector, Anthropic CEO Amodei has advocated slowing the pace of AI development to some extent and introducing stringent safety requirements.

Donald Trump, by contrast, is firmly opposed to restrictions and has given the technology a new name: “Superintelligence.” Previously, after Anthropic introduced safeguards, or “guardrails,” to prevent the military misuse of its technology, the US Department of Defense, known as the Pentagon, designated the company as a “supply chain risk.” The company is currently engaged in a legal battle over the designation.

Potential Impact on Future Technology Policy

Despite the strained relationship between the company and the administration at the policy-making level, the Trump-Amodei rivalry has also attracted attention in the US entertainment world, including on programmes such as Saturday Night Live (SNL). Nevertheless, some influential officials in the Trump administration are reportedly keen to maintain positive relations with Anthropic to protect commercial interests in the technology sector.

Market analysts believe this high-profile dinner bringing together Silicon Valley and Washington could have far-reaching implications for future AI development, commercial freedom and US government policy decisions.

//DBTech/BMT/OR//

The main focus of the redesign is the app’s bottom navigation bar. According to reports from app trackers, Meta plans to add its artificial intelligence feature, Meta AI, as a permanent option on the navigation bar.

Previously, Meta AI appeared as a floating icon just above the message-compose button. Under the new design, it will become part of the app’s main navigation menu.

Modernised Design and Interface Changes

As part of the update, the new Android navigation bar is being redesigned with a slightly oval shape and a semi-transparent appearance. The new layout bears a strong resemblance to the messaging navigation bar in Apple’s iOS operating system.

However, rather than adopting the fully transparent design used in iOS, WhatsApp’s interface maintains a degree of visual balance to avoid distracting users. Android users will be able to access the redesigned interface in both light and dark modes.

Beta Testing and Rollout Plans

The design changes were first spotted in WhatsApp’s experimental Android beta version 2.26.38.10. Currently, a limited number of users enrolled in the beta programme are getting the opportunity to try out the new interface.

Meta may need several more weeks to finalise the feature’s details before officially rolling it out to general users through the Google Play Store.

//DBTech/BMT/OR//

The company confirmed the attack to local media on Sunday, September 27.

Kyivstar said there were no casualties despite the direct strike on its headquarters. Meanwhile, Russia’s Defence Ministry claimed that a key data centre belonging to Vodafone Ukraine, the country’s second-largest mobile network operator, had also been hit.

Civilian Casualties and Disrupted Communications

Ukrainian President Volodymyr Zelenskyy said that two consecutive missile strikes hit a data centre in Kyiv on Sunday, injuring three civilians, including two children.

According to Ukraine’s Ministry of Digital Transformation, a series of attacks targeting internet service providers’ infrastructure over the past week has left approximately 100,000 households in Kyiv and surrounding areas without electricity and network connectivity.

Strategic Objectives and Response

Ukrainian Foreign Minister Andrii Sybiha wrote on social media platform X that Russia was strategically destroying data centres, primarily to disrupt the broadcasting of advance warnings about incoming missile and drone attacks.

He said that interrupting the flow of life-saving information during emergencies had become Moscow’s new military strategy. In this new phase of the conflict, Russian forces are reportedly intensifying efforts to cripple Ukraine’s information technology and communications sectors, alongside targeting military objectives.

//DBTech/BMT/OR//

The new device is expected to launch globally this autumn, potentially alongside the official debut of Apple’s new voice assistant, Siri AI.

Design and Colour Variants

According to leaked information, the HomePod mini 2 will retain much of the original model’s exterior design, with no major changes expected. However, it may offer a wider range of colour options, with the device reportedly available in five attractive shades: green, pink, light blue, white and black.

The first-generation model, launched in October 2020, was available in yellow, orange and dark blue, among other colours. The new version is also expected to feature a faster, more advanced chip or processor to support AI-powered features and execute voice commands more quickly.

Smart Home Devices and Apple’s Broader Plans

Alongside the HomePod mini 2, Apple is reportedly planning to introduce several new devices to strengthen its smart home ecosystem.

According to technology industry reports, the company may unveil a new Apple TV streaming box and a dedicated smart home hub featuring a display, alongside the smart speaker.

After a lengthy gap, Apple’s affordable smart device is expected by analysts to bring a new dimension to home entertainment and communication in the era of artificial intelligence.

//DBTech/BMT/OR//

After the fraud was detected, approximately €53 million was recovered with the assistance of international investigative agencies, while the remaining €36 million is still missing.

The incident has emerged as one of the largest technology-driven banking fraud cases involving the use of AI in recent times.

Fraudulent Tactics and Fake Instructions

According to reports, the fraud scheme began in February and primarily targeted Paolo Molesini, then chairman of Fideuram. He received a fraudulent message impersonating the CEO of Intesa Sanpaolo, instructing him to assist with an urgent international transaction.

Shortly afterwards, the chairman received a phone call from someone posing as a senior partner at a prominent law firm. Using AI-powered voice-cloning technology, the fraudsters convincingly replicated the lawyer’s voice to carry out the deception.

Believing the instructions to be genuine, Chairman Molesini directed the bank’s finance department to transfer large sums of money in stages to several overseas accounts in China and Hong Kong.

Recovery Efforts and Cybersecurity Risks

After detecting unusual transactions, the bank alerted international agencies. A joint effort involving Italy, China and Portugal led to the recovery of part of the stolen funds. However, tracking the remaining missing money has become difficult, as it was transferred through various shell accounts, converted into cryptocurrency and laundered.

Although no bank employees have been accused in connection with the incident, prosecutors in Milan, Italy, have launched an investigation into a non-European national.

The case has reignited global debate over the need to strengthen institutional security measures to prevent the misuse of AI technology in the corporate and banking sectors.

//DBTech/BMT/OR//

The new version features direct integration with Microsoft Office applications, along with a tab-based interface.

The Home tab allows users to get answers through ordinary conversations. For more complex tasks, it can automatically route prompts to Copilot’s advanced mode, significantly speeding up data analysis and presentation creation.

Coding Without Programming and the New ‘Code’ Tab

Copilot’s new Code tab enables users to create digital tools using natural-language prompts, without requiring any coding or programming skills. By simply describing what they need, users can automatically generate applications, data-management dashboards and desktop widgets.

Jared Spataro, Microsoft’s head of workplace AI, said Copilot would analyse the needs of everyday users and independently organise the code in the most suitable way to build the desired technology tools.

‘Autopilot’ Feature and Enhanced Task Automation

The most significant changes have been introduced in the app’s Autopilot tab, which functions as an independent AI agent capable of carrying out tasks autonomously.

Without direct human intervention, Copilot will now be able to automatically respond to email threads, schedule business meetings, review official documents and complete necessary follow-ups with partners.

With these new additions, Microsoft aims to move AI beyond the confines of chatbots and accelerate everyday workflows.

//DBTech/BMT/OR//

The lawsuit was filed last year against ByteDance, TikTok’s Chinese parent company.

The settlement was reached just before the trial was scheduled to begin in court next week. It marks TikTok’s first settlement with a US state in a lawsuit involving social media addiction.

New Safety Features and Requirements

In addition to paying the settlement, TikTok will introduce several stricter safety measures to protect younger users. Under the new rules, daily screen time for minors will be capped at two hours, with parents able to reduce the limit further if they wish.

The company has also agreed to restrict late-night app use, eliminate endless scrolling and completely prohibit beauty filters and other artificial appearance-enhancement features for younger users.

Algorithm-driven personalised content feeds will also be disabled for children. Under the agreement, TikTok could face an additional $300 million penalty if it fails to implement these changes.

Broader Context and Growing Pressure

TikTok is facing similar lawsuits in around a dozen other US states. Recently, technology giant Meta reached a massive $18 billion settlement with 47 US states over broadly similar allegations.

Earlier, in August, TikTok separately settled a lawsuit with the US Department of Justice for $400 million over alleged violations of children’s data protection rules.

To navigate social and political pressure in international markets, TikTok has already established a separate board of directors under non-Chinese investors to oversee its US operations.

//DBTech/BMT/OR//

Although this is the second phone in Motorola’s Signature series, the company has introduced an interesting change in its naming convention. The new flagship has been named Signature 27 based on the final two digits of the coming year. To compete with other top flagship smartphones on the market, it features Qualcomm’s latest and most powerful ‘Snapdragon 8 Elite Extreme Gen 6’ processor.

Advanced Cooling Technology and Artificial Intelligence

To prevent the device from overheating during demanding smartphone tasks or extended gaming sessions, Motorola has equipped it with its specialised ‘ArcticMesh’ cooling technology. The new cooling system features a large vapour chamber that is around 40 percent larger than the previous version.

As a result, the handset is expected to maintain stable performance even during complex AI-based tasks or intense processing workloads while keeping temperatures under control. The combination of the processor’s high performance and the improved cooling technology could make the phone particularly appealing to gamers and power users.

200-Megapixel Camera and a New World of Sound

The Motorola Signature 27 is set to offer technology enthusiasts a new photography experience. Its rear main camera uses a 50-megapixel Sony LYT-910 sensor with 3.5-degree optical image stabilisation (OIS).

The biggest surprise, however, is its powerful 200-megapixel periscope telephoto camera. The lens is designed to deliver clear images of distant subjects and impressive zoom capabilities.

Motorola has also signed a long-term agreement with renowned audio technology company ‘Bang & Olufsen’. As the first product of the partnership, the new phone will feature ‘Audio by B&O’ technology, designed to take music and video playback to another level.

Attractive Design and Long-Term Support

The premium handset also stands apart from conventional smartphones in terms of design. It will be available in two sophisticated colours—‘Pantone Coal Smoke’ and ‘Pantone Capulet Olive’. Both feature a distinctive textured finish on the rear panel, combining visual appeal with a secure grip that helps prevent the phone from slipping out of the hand.

Motorola has also offered customers a major assurance regarding software security. The company has announced seven consecutive years of operating system updates for the Signature 27.

Although the phone is expected to reach global markets toward the end of this year, customers will have to wait a little longer for the official launch date, complete specifications and final pricing.

//DBTech/BMT/OR//

From Samsung engineers to top US cybersecurity officials and ordinary users, the list of people facing legal and physical risks after sharing confidential or inaccurate information with ChatGPT continues to grow.

Even OpenAI CEO Sam Altman has acknowledged that conversations and information shared through the chatbot are not completely private or legally protected. There remains a risk that such information could be disclosed in response to a court order or for security-related reasons.

Although AI can be useful for research, coding and preparing various drafts, using ChatGPT in at least five sensitive areas can be extremely risky.

Legal Advice and Court Documents

Relying on ChatGPT in legal matters can lead to serious trouble. More than 2,000 legal cases worldwide have reportedly involved instances in which false information and fabricated references generated by ChatGPT were used.

Not only ordinary users but also professional lawyers have faced hefty fines after submitting court documents containing fabricated information generated by AI.

OpenAI has already introduced restrictions in its policies on providing legal advice without licensed professionals. Moreover, while communications with a lawyer may be protected by legal privilege, words entered into ChatGPT could potentially be subpoenaed by a court. Therefore, apart from learning general legal principles, relying on AI for documents or complex matters related to one’s own case can be extremely risky.

As a Substitute for a Therapist or Mental Health Professional

Millions of people talk to ChatGPT every week about depression and complex psychological problems. However, a chatbot can never replace an experienced therapist or counsellor.

Research has found that AI tends to support users’ existing thoughts in a flattering or agreeable manner. As a result, it could potentially cause serious harm to someone experiencing psychological distress or confusion. OpenAI has even faced lawsuits following tragic incidents in which AI was allegedly involved in harmful encouragement, including suicide.

For this reason, several countries are moving to legally restrict the use of AI for therapy or decision-making in mental health contexts. Treating ChatGPT as a doctor or therapist when dealing with personal emotional struggles or a mental health crisis is not safe.

Confidential and Institutional Information

Uploading sensitive workplace documents or confidential code to ChatGPT could potentially lead to serious professional consequences, including job loss. Recently, a senior US cybersecurity official caused controversy after uploading confidential government files to ChatGPT.

Similarly, after Samsung engineers used the company’s internal source code with ChatGPT, the company reportedly banned the use of AI tools altogether. Even if a user disables permission for their data to be used, information may still be stored on external AI servers and could potentially be exposed through hacking or disclosed during legal proceedings. Therefore, confidential corporate agreements or customer data should never be uploaded to an AI chatbot.

AI Prescriptions Instead of a Doctor

Seeking medical advice or medication recommendations from ChatGPT for physical illnesses could have life-threatening consequences. AI has not been proven sufficiently reliable for assessing a patient’s medical history and determining the correct medication or dosage.

Recently, an elderly man reportedly asked ChatGPT for a substitute for salt in his diet and was advised to consume bromide. He purchased the toxic chemical and consumed it for three months, eventually becoming seriously ill and requiring hospitalisation.

Medical organisations and OpenAI itself have clearly stated that following AI-generated medical advice without consulting an experienced doctor can be extremely dangerous.

Passwords and Personal Banking Information

You should never enter your passwords, national identification number, bank account details or passport information into ChatGPT. In the past, numerous private ChatGPT chat histories were accidentally exposed through search engines.

Moreover, due to a copyright-related lawsuit, deleted user chats reportedly had to be retained on servers for an extended period under a court order. ChatGPT can also retain information provided by users in its memory, potentially allowing it to surface inadvertently in a future conversation. Therefore, to protect against cyber fraud and identity theft, users should completely avoid sharing sensitive personal information with ChatGPT.

//DBTech/BMT/OR//

The platform is also introducing a feature that allows creators to publish episodic content in the Shorts format, enabling videos to be arranged sequentially. The new option is expected to create a different experience for fans of microdramas and short-form series.

GIFs and Voice Input in Comments

To encourage greater social interaction, YouTube’s mobile app will introduce a searchable GIF library in the comments section, allowing users to express themselves through animated images instead of relying solely on text.

On smart TVs, users will also be able to post comments directly by speaking, using text-to-speech technology.

Meanwhile, YouTube has strengthened its ‘Ask YouTube’ and shopping features. Users will be able to ask the chatbot about the prices and comparative information of clothing and other products shown or worn in videos.

New Technology for Music and Live Dubbing

Alongside video, YouTube Music and its podcast service are also receiving generative AI features such as ‘Ask Music’ and ‘Podcast Lineup’. These tools will allow users to obtain information about musicians and listen to automatically generated preview summaries of their favourite weekly podcasts.

For international users, the platform is also planning to introduce real-time automated dubbing during live video broadcasts in early 2027.

Analysts believe these wide-ranging technological developments will further strengthen YouTube’s position in content creation, distribution and communication.

//DBTech/BMT/OR//

Powered by Meta’s new AI assistant ‘Muse’, the glasses allow users to answer phone calls, listen to music and podcasts, and translate languages in real time. Pre-orders for the glasses will open on October 13 at a price of $349.

Lightweight Design and Battery Performance

Developed through a collaboration with renowned eyewear manufacturer EssilorLuxottica, the glasses are considerably slimmer than previous versions because they have no camera and weigh just 43 grams. They offer up to 12 hours of continuous battery life on a single charge, which can be extended to as much as 48 hours with the charging case.

Alongside two new classic frame styles, ‘Clubmaster’ and ‘Burbank’, the glasses will be available in a total of 23 colour combinations. Prescription lenses can also be fitted according to the user’s needs.

Gen 3 Version with Camera and Other Updates

Alongside the audio-only glasses, Meta also announced the next-generation ‘Ray-Ban Meta (Gen 3)’ glasses with a camera. They feature a 12-megapixel camera capable of recording 3K Ultra HD video, up to nine hours of battery life and an enhanced audio system with six microphones.

The company also announced a new partnership with pop star Lisa and plans to expand the glasses into markets worldwide. To strengthen data protection, the new devices feature ‘Private Processing’. With this strategy, Meta is seeking to establish a stronger position in the audio wearable technology market, particularly as a competitor to Apple’s AirPods.

//DBTech/BMT/OR//

The constant noise can be enough to disrupt concentration during work or in the middle of an intense gaming session. However, excessive fan noise is not always a sign of a serious problem. In most cases, particularly in gaming or high-performance laptops, the fans need to spin rapidly to expel the excess heat generated by the processor (CPU) and graphics card (GPU).

In other words, the fans are making that noise to keep the laptop safe. However, in some cases, excessive noise can also raise concerns about the device’s condition and security, which can often be addressed with a little attention.

Why Does Your Laptop Suddenly Roar Like a Jet?

When a laptop is subjected to workloads beyond its capabilities, its cooling fans begin spinning at maximum speed. While running modern, graphics-intensive games such as Crimson Desert or performing demanding tasks such as 4K video rendering, the processor and graphics processing unit can become extremely hot. If this excess heat is not removed quickly, internal components could be damaged.

High-performance gaming laptops such as the ASUS ROG Zephyrus G14 feature a ‘Turbo Mode’ designed to deliver maximum performance. While enabling this mode can make games run exceptionally well, fan noise can exceed 50 decibels, making it considerably louder than a typical room environment.

Fan noise can also increase significantly when numerous unnecessary software programs continue running in the background or when the laptop’s ventilation pathways become blocked.

Is Excessive Fan Noise Harmful to Your Laptop?

If the fan becomes loud while playing a demanding game such as Cyberpunk 2077 or performing complex graphics-intensive tasks, there is generally no reason to panic. It does not necessarily indicate that the laptop is being damaged; rather, the fans are doing their job properly by keeping the internal components cool.

The real concern arises when you are doing something relatively light—such as switching between just one or two browser tabs—yet the laptop fan is roaring as loudly as an aircraft taking off.

If the fan continues running at maximum speed despite there being little workload, it may indicate that dust has accumulated inside the laptop and is blocking airflow. In such cases, the device should be taken to a qualified repair shop, where specialised equipment or compressed air can be used to clean out the accumulated dust.

On the other hand, if the fan starts producing unusual clicking, rattling or grinding noises, the situation may be more serious. This could indicate that the fan’s internal bearings or rotating components have been damaged. In such circumstances, the laptop could overheat severely and suddenly shut down. If you hear such noises, you should promptly seek assistance from an experienced technician or the manufacturer’s authorised service centre.

Simple Ways to Reduce Laptop Fan Noise

There are several simple ways to reduce unwanted noise from a laptop. First, Windows laptops allow users to change their power settings. Go to Settings, select ‘System’, and then open ‘Power & battery’. Under ‘Power mode’, select ‘Best power efficiency’. This reduces the system’s power consumption, which can lower both fan speed and noise.

You can also control fan noise by enabling a ‘Quiet Mode’ through the laptop manufacturer’s own software, such as ASUS’s ‘Armoury Crate’.

To reduce excessive fan activity while gaming, lowering some in-game settings can be highly effective. Limiting the game’s frame rate to 60 frames per second and disabling demanding visual features such as ‘ray tracing’ can significantly reduce the load on the graphics card. As a result, the fans can run more quietly.

It is also important to pay attention to where you use the laptop. Placing it on a soft mattress, bed or lap cushion can completely block the ventilation openings. Therefore, always use the laptop on a desk or another hard, flat surface so that the ventilation openings underneath remain unobstructed and the fans do not have to work unnecessarily hard.

//DBTech/BMT/OR//

On Tuesday, AI research company Anthropic unveiled its new artificial intelligence model, ‘Claude Opus 5.5’. Meanwhile, OpenAI, the creator of ChatGPT, introduced two relatively low-cost models, ‘Sol’ and ‘Luna’, under its latest ‘GPT-6’ series.

Anthropic’s ‘Claude Opus 5.5’

Anthropic said its new ‘Claude Opus 5.5’ model can be operated at around 40 percent lower cost than the previous ‘Opus 5’ version. Despite the lower cost, it is capable of delivering performance comparable to the company’s top-tier ‘Fable 5.1’ model.

The model is priced at $4 per 1 million input tokens and $20 per 1 million output tokens—around 20 percent lower than the pricing of Opus 5.

In an international benchmark for software development and coding, the new Opus 5.5 outperformed rival OpenAI’s ‘GPT-5.6 Sol’ model, while costing roughly one-third as much to operate.

The model is already available through Amazon Web Services (AWS), Google Cloud and Microsoft Azure.

Anthropic has also announced plans to launch ‘Claude Sonnet 5.5’ and ‘Claude Haiku 5.5’ within the next few weeks. Additional capabilities are also being developed for authorised cybersecurity and life-sciences researchers.

OpenAI’s ‘GPT-6’ Series: Sol and Luna

Earlier this month, OpenAI unveiled ‘Astra’, the fastest and most powerful model in its GPT-6 series. It has now introduced two lower-cost models designed for professional work and automation.

The company said ‘GPT-6 Sol’ and ‘GPT-6 Luna’ are particularly suited to professional tasks, coding, automation technologies and computer-based autonomous operations.

Processing and inference improvements have significantly reduced the cost of the two models. GPT-6 Sol will cost $2 per 1 million input tokens and $10 per 1 million output tokens—50 percent lower even than the promotional pricing of the previous GPT-5.6 Sol.

The more affordable ‘GPT-6 Luna’ will cost $0.10 per 1 million input tokens and $0.50 per 1 million output tokens.

OpenAI said Astra will remain its flagship model for complex tasks, while Sol and Luna have been developed to retain similar reasoning capabilities and accuracy at lower cost.

Safety and Risk Concerns

Alongside advances in AI technology, fresh concerns have emerged over AI safety and the potential risks associated with controlling increasingly capable systems.

Anthropic said its Opus 5.5 model underwent testing by independent AI safety research groups, including ‘Frontier Design’ and ‘METR’, before its release. In internal safety testing, it made around 85 percent fewer attempts to cross designated software boundaries or containment limits than the previous ‘Opus 5’ or ‘Mythos 5.1’ models.

Earlier this month, Anthropic CEO Dario Amodei called for a slight slowdown in the pace of new technology releases in the interest of safety.

OpenAI, meanwhile, has warned that its powerful ‘Astra’ model may occasionally attempt to evade human oversight. Such behaviour has come under heightened scrutiny following recent incidents in which various AI agents reportedly attempted to access other organisations’ systems without authorisation.

//DBTech/BMT/OR//

After a successful 15-year tenure as CEO, Cook stepped down on September 1 and took over as executive chairman. During his tenure, Apple’s share price rose by nearly 2,000 percent.

While Steve Jobs built Apple into an iconic brand, Cook created one of the world’s most reliable and profitable commercial foundations.

Product Diversification and Ecosystem Expansion

Although Cook was unable to introduce another revolutionary device on the scale of the iPhone, he captured huge markets with wearable products such as the Apple Watch and AirPods. His greater achievement, however, was the extraordinary expansion of Apple’s services business around the iPhone, including Apple Music, iCloud and Apple Pay, which generated $109 billion in revenue last year alone.

Cook also faced criticism over the early confusion surrounding Apple Maps and the cancellation of the company’s self-driving car project after a decade of development. At the same time, he demonstrated far-reaching leadership in protecting customer privacy and navigating US tariff policies while maintaining diplomatic relations with the Trump administration.

Next Leadership and Future Direction

Leaving behind a massive $147 billion cash reserve and a stable business, Cook will now serve as Apple’s executive chairman. Meanwhile, veteran hardware executive John Ternus has taken over as Apple’s new CEO.

Recently appearing on stage to unveil Apple’s foldable phone, the ‘iPhone Duo’, Ternus made it clear that Apple will seek to reshape the world in new ways while building on the strong foundation created by his predecessor.

//DBTech/BMT/OR//

The chipsets feature a new ‘Sensing Hub’ capable of running small AI models with up to 200 million parameters locally, without an internet or cloud connection.

As a result, smartphones will be able to recognise users’ voices and take notes, suggest tasks based on usage patterns, and run automated voice-based AI directly on the device with greater ease.

Extreme Version, Camera and Audio Features

The more powerful ‘Extreme’ version can process a 30-billion-parameter ‘Mixture-of-Experts’ AI model directly on the device, making the process faster and more cost-efficient.

For photography and videography, the processor offers pixel-level control, enhanced stabilisation, 8K video recording at 60 frames per second and 4K ultra-slow-motion video at 240 frames per second. It also features Qualcomm’s proprietary ‘Voice Bubble’ technology to eliminate unwanted background noise during voice calls.

First Smartphone and a Shift in Market Thinking

At the event, smartphone maker Motorola officially announced the first use of the new technology, saying its upcoming ‘Motorola Signature 27’ flagship phone will be powered by the Snapdragon 8 Elite Extreme Gen 6 processor.

Qualcomm believes users can get the full benefits of AI directly from their smartphones rather than relying on separate AI gadgets.

The growing shift toward this approach among Apple and brands popular with Gen Z suggests that the future mobile market is increasingly moving toward fully on-device AI technology.

//DBTech/BMT/OR//

On Tuesday, September 22, he made the remarks as chief guest at the global dialogue titled “Empowering Teachers in Low-Resource Classrooms to Co-Design EdTech for Student-Centered Learning” at Education House in New York, held on the sidelines of the 81st session of the United Nations General Assembly. According to the organizers, the session focused on teacher-led development of education technology for low-resource classrooms and the effective use of AI to support student-centred learning.

The state minister said, “The future of education is technology-driven. However, teachers must be equipped with the necessary skills to ensure the best possible use of technology. For this reason, training on AI and modern education technology will be arranged for primary school teachers.”

He said, “Technology can never replace teachers; rather, it is an effective tool for enhancing their capabilities. Particularly in low-resource classrooms, it is necessary to ensure the appropriate use of technology to make learning opportunities more effective and inclusive for students.”

Bobby Hajjaj said, “We want our primary school teachers not only to use technology, but also to play an innovative role in determining how technology can be used according to students’ needs. Student-centred education can be ensured by bringing teachers’ experience and the potential of technology together.”

The state minister further said that alongside expanding the use of digital technology in primary education, equal importance must be given to teacher training, classroom realities, student learning outcomes and local needs. He said Bangladesh would use international experience and support from partners to undertake effective technology-based initiatives in primary education.

The panel discussion featured Dr Ahmed Mohamed, Minister of State for Education, Higher Education and Skills Development of the Maldives; Alex Wong, Senior Adviser at the Strategic Engagement and Initiatives Division in the Office of the Secretary-General of the International Telecommunication Union (ITU); and Dr Borhene Chakroun, Director of the Division for Policies and Lifelong Learning Systems at UNESCO. Education House also listed them among the speakers for the session.

Earlier this month, work was also undertaken at the government level on an ICT-AI competency framework for primary teachers. On September 8, a workshop on the “ICT-AI Competency Framework for Primary Teachers” was held at the Primary Teachers’ Training Institute in Mirpur with technical support from UNESCO Bangladesh. Bobby Hajjaj attended the event as chief guest.

Later, the state minister attended as chief guest an event organized by the Bangladesh Consulate General in New York titled “From Classroom to Economy: Modernizing Bangladesh’s Education and Skills—Turning Learning into Jobs, Enterprise and Economic Strength.”

Speaking at the event, he said an effective connection must be established between classroom education and real-life skills. The future generation must be prepared in a way that enables them to apply their knowledge and skills to employment, entrepreneurship and the country’s economic development.

The event discussed various aspects of modernizing Bangladesh’s education and skills development system, promoting employment-oriented education and establishing stronger links between education and economic activities.
//DBTech/MTM/SME/MIST//

The feature keeps a detailed record or log of the places you visit every day. While it can be useful when trying to remember the name of a restaurant you visited in the past, many people consider it a significant intrusion into their privacy.

For some users, having their location continuously recorded in the background can feel like constant surveillance. There are also potential security risks if such information falls into the wrong hands.

Google recently introduced a system that stores users’ Location History directly on their personal devices instead of keeping it on its cloud servers. Even so, many users may still prefer to disable location tracking altogether. Fortunately, turning off Google Maps’ Location History or Timeline feature is relatively simple.

How to Stop Tracking from Your Smartphone

Google Maps Timeline or location tracking can be disabled through the My Activity section of a Google Account or directly through the mobile app. However, it is important to remember that the setting applies to a specific device.

This means that if you use Google Maps on a tablet or multiple smartphones, you will need to follow these steps separately on each device. If you later decide to turn the feature back on, you can follow the same process and enable it at the final step.

To turn off Location History, first open the Google Maps app on your smartphone and tap your profile picture on the right side of the search bar. From the menu that appears, select Your Timeline. On the next screen, tap the three-dot icon in the upper-right corner and select Location & Privacy Settings. From there, go to Timeline, where you will find a drop-down menu.

If you want to stop future tracking while keeping the data already stored, select Turn Off. If you want to stop tracking and permanently delete all previously stored history, choose Turn Off and Delete Activity.

Automatic Data Deletion and Other Privacy Options

If you want to keep recent travel or location information for a while but have older data deleted regularly, Google also offers an automatic deletion feature. Go to the Auto-delete option near the bottom of the same settings menu. From the drop-down list, you can choose to have data older than three, 18 or 36 months automatically deleted.

You can also tap the cloud-backup icon at the top of the main Timeline page to enable or disable an encrypted backup of your device’s Timeline data to your Google Account.

To edit or delete the places visited on a specific day, you can use the pencil icon below. If you want to remove information from a particular period all at once, the Delete range of Timeline data option allows you to clear the selected data and maintain greater control over your location privacy.

//DBTech/BMT/OR//

During trading on the US stock market on Monday, AMD’s share price jumped 9.6 percent to a record $613.31.

With the milestone, the California-based company became the fourth US chipmaker to join the exclusive $1 trillion club, following Nvidia, Broadcom and Micron. AMD’s stock valuation has surged by around 185 percent in 2026.

Business Expansion and New Strategy

According to market analysts, investors are once again pouring substantial capital into AI-focused technology stocks despite a global economic slowdown and geopolitical tensions between the United States and Iran.

As Nvidia’s leading rival in the AI-driven graphics processing unit (GPU) market, AMD has moved beyond relying solely on chip sales. The company has begun offering all-in-one server systems that integrate networking equipment and hardware, strengthening its position in the rapidly expanding AI infrastructure market.

At the same time, AMD has strengthened its business foundation by capturing a significant share of the server-processor market from rival Intel.

Broad Semiconductor Market Rally

AMD’s achievement has also boosted the broader semiconductor industry. On Monday, the Philadelphia Semiconductor Index rose 2.7 percent, reaching its highest level in a month. AMD’s major rival Intel saw its shares jump 11.8 percent, while Qualcomm gained 4.5 percent.

Although investors’ exceptionally high expectations were not fully met by the quarterly earnings outlook announced last month, continued global demand for chips to build AI infrastructure is providing fresh momentum to the technology market.

//DBTech/BMT/OR//

As Google’s European headquarters are located in Dublin, the DPC conducted the investigation on behalf of the EU. In addition to the fine, Google has been given a final deadline of six months to bring its location-data processing practices fully into compliance with the European GDPR framework.

Findings and Regulatory Observations

The lengthy investigation, launched in 2020 following complaints from several consumer-protection organisations, examined three specific Google features covering the period from May 2018 to February 2020. These were Web & App Activity, Location History and Android’s Location Accuracy.

According to the DPC’s findings, Google failed to maintain sufficient transparency about how users’ location information was being used. In particular, the regulator found that retaining the data for longer than necessary for purposes such as advertising and assessing users’ interests, without adequate consent, violated GDPR principles. The fine ranks as the fourth-largest penalty imposed by the DPC since the GDPR came into effect.

Google’s Response and Wider Context

Responding to the ruling, Google said the legal issues in question relate to its older policies, which have since been revised. The company said it had made significant changes to its data-protection practices since 2019 and had given users greater control over their information, including features that automatically delete data.

This is not Google’s first legal challenge in the European market. The technology giant previously faced a $4.7 billion antitrust case in 2018 and a $1 billion fine over allegations that it gave preferential treatment to its own services in search results.

//DBTech/BMT/OR//

Under the agreement, Deezer will provide its licensed music catalogue and streaming infrastructure, while Winamp will operate under its own brand name and manage customer services. The new app will allow users to access local files stored on their devices, podcasts and internet radio, while also offering millions of songs through a monthly subscription.

Deezer signed the agreement with Winamp as part of a new business-expansion strategy following a decline in subscribers and revenue during the first half of 2026.

Notably, Winamp’s original desktop version still has around 40 million active users worldwide. An earlier attempt to revive Winamp as a streaming platform in 2018 failed because of complications surrounding commercial licensing.

Later, in September 2024, the software’s source code was made available on GitHub. However, it was removed within weeks amid legal disputes and concerns following code leaks involving other major companies.

Although the company has expressed optimism about the planned new version of Winamp, it has yet to disclose details about its subscription pricing or operating-system compatibility. It has, however, confirmed that the service will feature theme and interface customisation, social connectivity and advanced technology for easily discovering music.

After nearly three decades of changing ownership—including AOL and Radionomy—this will be the first premium subscription service in Winamp’s history.

//DBTech/BMT/OR//

A closer look at the technology landscape reveals a vast range of completely free and legal entertainment options. From channel surfing like traditional cable TV to selecting movies from a personal watchlist, these free services offer a wide variety of viewing options.

Since these platforms provide content at no cost, viewers naturally have to put up with occasional advertisements. With a little smart use of a smart TV, however, free entertainment can become an excellent companion during leisure time.

Free Channels Built into Your TV

You do not have to look far to begin watching TV for free. Almost every smart TV currently on the market comes with certain free services pre-installed, commonly known as Free Ad-Supported Streaming Television (FAST) platforms. Along with on-demand movies, these services offer a variety of channels that broadcast around the clock, much like traditional cable television.

Services such as LG Channels on LG TVs, Samsung TV Plus on Samsung devices and Google TV Freeplay are already integrated into their respective television platforms.

While these services may not offer the latest blockbuster movies or popular live sports broadcasts, they provide plenty of game shows, lifestyle programmes, cooking shows, news and movies across a range of genres.

Two Popular Ad-Supported Streaming Platforms

Those looking to expand beyond their TV’s built-in apps can download popular free services such as Tubi and Pluto TV from their app stores.

Tubi allows users to watch movies and dramas freely without paying a subscription fee—or even creating an account or signing up. Alongside its extensive movie library, the platform offers content across various categories as well as a diverse selection of Tubi Originals.

Pluto TV, meanwhile, is particularly known for its extensive collection of live channels. It offers an experience reminiscent of traditional cable television. Crime-series fans can find dedicated channels that continuously broadcast the same shows, while there are also channels featuring anime, local news and music.

World-Class Films with a Library Card

For people who have library memberships abroad, two services called Hoopla and Kanopy offer another excellent gateway to entertainment. Users can access these apps for free with valid library-card credentials from participating libraries.

Hoopla offers a wide selection of dramas, movies and comics from major broadcasters and content providers around the world. Kanopy, on the other hand, focuses less on mainstream commercial films and more on documentaries, non-fiction productions and independent and international cinema. Neither platform interrupts viewing with advertisements.

YouTube’s Official Free Movie Section

For many people, YouTube is synonymous with ordinary videos uploaded by users. However, the platform also offers a vast collection of official free movies through legitimate, copyright-compliant channels.

By going to the Movies & TV section from YouTube’s left-hand menu and selecting the Free category, users can access hundreds of movies and series at no cost. Official YouTube channels operated by major production houses also offer a wide range of free content.

From the Warner Bros. Classics channel to hugely popular cartoons from earlier eras and cult classic films, a variety of entertainment can be easily enjoyed through the YouTube app on a smart TV.

Amazon Prime and Sling Open New Possibilities

Users can also watch a substantial amount of free content through the Amazon Prime Video app without paying. Although such content was previously offered through a separate app, Amazon has now incorporated a large free-to-watch library directly into the Prime Video app.

Alongside paid and subscription-based content, the app features a dedicated free category offering both on-demand movies and live channels.

Another platform, Sling Freestream, designed along the lines of US cable television, allows users to watch various news and reality-TV channels without even creating an account. Creating a free account provides additional benefits, including the ability to record or DVR around 10 hours of favourite programmes.

//DBTech/BMT/OR//

According to the IFR’s criteria, a robot is considered humanoid if it has a human-like external appearance and can operate autonomously in environments designed for humans. However, having legs is not a mandatory requirement.

Although headlines regularly highlight billions of dollars in investment and impressive demonstrations by major technology companies, the number of humanoid robots remains extremely small compared with the total number of robots in use worldwide.

Applications and Limitations

IFR Secretary General Susanne Bieller said a significant portion of the humanoid robots sold last year are not directly involved in productive work. Instead, many are being used in research laboratories and technology companies to collect data that can help train and improve artificial intelligence (AI) models.

Automakers were among the early adopters of the technology, but the number of humanoid robots currently operating in factories remains in the single or low double digits.

By comparison, around 542,000 conventional industrial robots and nearly 200,000 service robots were added worldwide in 2024, highlighting how early humanoid robots remain in their development and adoption cycle.

Future Potential and Market Trajectory

Despite their limited initial sales, market analysts predict that the humanoid robotics sector could experience explosive growth in the coming years. China, the world’s largest market for industrial robots, is emerging as a particularly promising market for the technology.

According to estimates by Bank of America Global Research, around 90,000 humanoid robots could be delivered worldwide in 2026, with annual global sales potentially reaching approximately 1.2 million units by 2030.

Beyond consumer and military applications, the commercial use of such robots is expected to transform industrial production, labor markets and the very definition of automation technology worldwide in the years ahead.

//DBTech/BMT/OR//

The first step can be Windows’ built-in security feature, Microsoft Defender. In addition to its standard Quick Scan and Full Scan options, users can restart the PC and run an Offline Scan. This uses the Windows Recovery Environment to detect and remove sophisticated malware that may be hiding within the system.

Scanners and Other Tools

For additional protection, it is advisable to run a second check using an on-demand scanner such as Malwarebytes. Since it does not need to remain active in the background at all times, it can be used whenever necessary to scan the PC and detect malware that may have evaded Defender.

For removing unwanted adware, AdwCleaner can be useful, while VirusTotal is effective for checking suspicious files or web links. In cases of more complicated infections, users can boot the PC into Safe Mode and run a scan. Malware can also be removed using Windows’ built-in Malicious Software Removal Tool (MSRT), which can be launched by pressing Win + R and entering mrt.

Paid Antivirus and Preventive Measures

For most ordinary users, automatic Windows Defender protection combined with an on-demand free scanner is generally sufficient, meaning there is little need to spend extra money on third-party antivirus software. Such third-party programs also frequently generate annoying notifications prompting users to upgrade.

However, it is important to remember that no security software can guarantee 100 percent protection against cyber threats. No scanner can protect users from account hacking caused by weak passwords, online scams, or mistakes such as clicking phishing links. Therefore, rather than relying solely on malware scans, avoiding suspicious emails and unfamiliar websites remains one of the best ways to protect a PC.

//DBTech/BMT/OR//

According to data from Downdetector, the technical issue began at around 8:55 p.m. US time. At the peak of the outage, more than 17,000 Facebook users and over 5,000 Instagram users reported problems. A large number of users said they were unable to access the websites or refresh their home feeds on the mobile apps.

However, within about an hour of the disruption, around 10 p.m., the number of reports dropped sharply. This indicated that services had been restored for most users relatively quickly.

Meta did not immediately issue any official statement or comment on the cause of the temporary outage. It is worth noting that Downdetector’s figures are primarily based on user-submitted reports, meaning the actual number of affected users may have been somewhat different.

According to technology analysts, temporary disruptions of this kind can occur due to complications involving network infrastructure and Meta’s servers. Facebook and Instagram experienced a similar technical outage in July, when several thousand users temporarily lost access to the platforms.

As the technology sector continues to expand, maintaining the reliability of social media services has become a major challenge for Meta.

//DBTech/BMT/OR//

Citing internal documents from Volkswagen’s management board, the report said the new proposal is aimed at addressing a shortfall of around €700 million (approximately $803.8 million) in annual operating costs and implementing a broader restructuring plan. If approved, the cuts would come on top of Porsche’s previously announced workforce reduction program.

Major Blow to Porsche’s Overall Workforce

In July, Porsche management and employee representatives reached an agreement to cut an additional 5,000 jobs on top of 4,000 previously announced reductions. Under the plan, one in every five employees at the maker of the iconic ‘911’ model could be affected by job cuts by 2035.

If another 4,100 jobs are eliminated, the company’s workforce would face a major disruption. However, Volkswagen can only recommend the measure, and Porsche’s management is not legally obligated to implement it directly.

China Market Slump and Shift in EV Strategy

Volkswagen has reduced its profit target for the current year from 4.0–5.5 percent to a maximum of 1 percent. The parent group is facing substantial financial losses largely due to a writedown in the value of its stake in Porsche.

Porsche has been under increasing pressure as vehicle sales have fallen sharply in global markets, particularly in China. The company is also struggling to reassess its strategy after spending heavily on electric vehicle (EV) development. Porsche CEO Michael Leiters is facing the challenge of navigating the company through this difficult period.

Against this backdrop, Volkswagen is considering such drastic measures in an effort to protect Porsche from further financial losses and restore its competitiveness.

//DBTech/BMT/OR//

The US International Trade Commission (USITC) has formally launched an investigation following a complaint filed by California-based small audio technology company “BoomCloud 360.” If the allegations are proven in the commission’s proceedings, the companies could face restrictions on the import and sale of certain smartphones and audio devices in the US market.

Patent Dispute and BoomCloud 360’s Claims

In August this year, BoomCloud 360 filed a complaint with the USITC alleging infringement of three specific patents. The California-based startup primarily develops software for cloud-based sound optimization and three-dimensional audio experiences.

The company claims that the accused firms are using its patented audio processing technology in their devices without authorization. In its legal filing, BoomCloud 360 has sought a “Limited Exclusion Order” and an injunction. If granted, the orders could halt the commercial supply of devices that use the disputed technology.

However, the specific devices targeted by the complaint have not yet been officially disclosed.

Tech Giants’ Countermoves and What Lies Ahead

The three tech giants have already intensified their legal efforts to avoid the risk of major product restrictions. Apple, Samsung, and Google have filed petitions in court seeking a “declaratory judgment” against BoomCloud 360’s claims, asking the court to establish that the technologies they developed are legally valid.

The outcome of the USITC’s high-profile investigation and subsequent court proceedings will determine whether the popular technology products involved can continue to be sold in the United States or whether the dispute will be resolved through a legal settlement.

//DBTech/BMT/OR//

Staged under the central theme “Advancing the Agentic World,” the summit brought together industry executives, enterprise clients, software architects, and academic researchers. Discussions highlighted the expanding role of AI agents and intelligent systems in refining corporate decision-making, lifting cross-sector productivity, and unlocking structural economic opportunities. Huawei underlined that equitable AI benefits demand transparent governance and shared progress among governments, commercial enterprises, tech partners, and developers.

Delivering the keynote address, David Wang, Deputy Chairman of the Board and Rotating Chairman of Huawei, unveiled a suite of advanced AI computing and data infrastructure solutions engineered to satisfy intensive computational workloads.

The primary technological reveals included the Atlas 960E SuperPoD, Huawei’s next-generation scalable AI computing architecture; the enhanced TaiShan 950 SuperPoD, optimized to deliver robust parallel processing for cognitive enterprise workloads; and the OceanStor M900, an ultra-resilient data-storage platform built specifically for high-throughput AI model training and inference workflows. These additions establish resilient foundations for organizations handling expanding data volumes while accelerating practical AI adoption.

Wang stated that the global economy is stepping into an AI-defined era of deep structural transition. He emphasized that robust hardware foundations, open-source compatibility, and ecosystem synergies are non-negotiable prerequisites to commercialize frontier technologies across key verticals, including telecommunications, digital banking, industrial manufacturing, clean power, logistics, and education. The event concluded with technical tracks examining digital workforce upskilling and next-generation connectivity standards.
//DBTech/FTP/DPO/BPO//

In this 14th test mission, the rocket will rise to an altitude of around 171 miles above the Earth’s surface and complete six orbits of the Earth over a 10-hour period. While all previous tests were suborbital, SpaceX considers this orbital mission a major milestone toward demonstrating the spacecraft’s overall reusability and preparing for its first commercial flights.

‘Starlink V3’ Satellites and a Leap in Speed

The most important aspect of the mission is the deployment of 26 next-generation “Starlink V3” internet satellites in space for commercial customer service for the first time. While previous tests carried satellites into space solely to demonstrate capabilities, this will be the first time they are deployed in orbit specifically to provide services directly to customers.

The V3 satellites will have a downlink capacity of 1 terabit per second, making them 10 times faster than the previous “V2” version. Their uplink speed will also reach 160 Gbps, nearly 22 times higher than that of the previous generation.

Landing Strategy and Live Broadcast

However, the flight will not feature the complex maneuver of using the launch tower’s mechanical arms to catch the rocket’s booster. A few minutes after liftoff, Starship’s “Super Heavy” booster will fall into the Gulf of Mexico.

Meanwhile, after completing its scheduled tasks in space, the main Starship spacecraft will land in the Pacific Ocean near the coast of Chile. SpaceX will livestream the entire launch process through its social media account on “X.”

//DBTech/BMT/OR//

According to a report by technology news site Engadget, CC has its own Google account, through which up to six members of a family can share selected information.

Users can decide which emails they want to share with CC. They can also forward individual messages when needed or set specific email addresses so that incoming messages are automatically routed to the CC account.

CC can create shared calendars and task lists that are accessible to participating family members. It can also send daily summaries covering upcoming events, scheduled activities and tasks that need attention.

The platform is designed to assist with practical and administrative tasks such as organising documents, filling out registration forms and preparing shopping lists. However, CC is specifically designed to seek user permission before taking action on their behalf.

Google initially launched the tool as an experiment in December last year. Earlier this year, the project was renamed and incorporated into the company’s Daily Briefing feature. Following requests from early testers, Google is now reshaping CC as a family-focused organiser.

Users can add CC to selected Google Drive folders or share specific files with the AI agent. They can also communicate directly with CC through a dedicated Google Chat window.

Google has outlined several privacy and security restrictions in its FAQ. CC cannot independently add or remove family members, nor can it access or search members’ private inboxes.

Existing CC users will be invited through prompts to upgrade their accounts and try the new tools. Google is currently maintaining a waitlist for new users interested in outsourcing parts of their daily planning and organisational tasks to the agentic AI system.
//DBTech/Google Blog/SME/MIS//

September 19, 2026 (Saturday): The university’s Public Relations Department announced the new tentative date in a notice issued Friday night. Professor Dr. Kazi Delwar Hossain, chairman of CUET’s Undergraduate Admission Committee and dean of the Faculty of Electrical and Computer Engineering, confirmed the information.

According to university sources, the administration has already begun preliminary preparations to ensure the admission test is conducted smoothly. As part of the process, the formation of the admission committee is currently underway and will be announced after receiving formal approval from the Academic Council.

Meanwhile, as part of the continuing engineering education admission process, Khulna University of Engineering and Technology (KUET) and Rajshahi University of Engineering and Technology (RUET) have already separately announced the schedules for their undergraduate admission tests for the upcoming 2026–27 academic year.

//DBTech/MAI/SME/OR//

In an announcement on Thursday, September 17, Amazon said the total value of employee benefits, including healthcare and education assistance, would exceed $32 per hour when combined with wages and other benefits.

Amazon is investing more than $1.5 billion in the latest wage and benefits programme, aimed at supporting and motivating its frontline workforce. The investment represents a relatively small portion of the company’s reported market value of around $2.68 trillion.

Alongside higher wages, Amazon is introducing Day 1 Financial, a low-cost banking service intended to help employees manage their finances. The company is also offering membership in First Tech Federal Credit Union.

Employees will receive a 20% discount on in-store purchases at Amazon-owned premium grocery chain Whole Foods Market and a 10% discount on online grocery orders. Amazon already provides benefits including free Prime membership, education assistance and healthcare.

The wage and benefits announcement, however, has drawn mixed reactions among Amazon employees. Some workers on online forums have described the $1 hourly increase as insufficient compared with current living costs.

Some employees have also responded critically to the Whole Foods discount. They argue that because Whole Foods products can be more expensive than those at conventional supermarkets, a 20% discount may still leave some purchases at prices comparable to those at other grocery stores.

//DBTech/techcrunch/BMT/IH//

According to the CEO of chipmaker SK Hynix, 2027 could be one of the worst years in the history of the technology industry in terms of chip supply, with the shortage potentially continuing until 2030.

According to a forecast by research firm Counterpoint Research, the global smartphone supply could fall 13.9 percent this year to 1.08 billion units due to the memory chip shortage, marking the largest annual decline in history.

For handsets priced at around $400, memory alone can account for as much as 60 percent of the total component cost, making the production of low-cost or entry-level smartphones increasingly unprofitable.

According to TrendForce data, contract prices for DRAM chips nearly doubled at the beginning of the year. Although the pace of price increases has now slowed somewhat to between 13 and 18 percent, supply conditions have yet to return to normal.

With limited capacity to stockpile large quantities of raw materials, small companies are adopting innovative approaches to survive the crisis. Finnish mobile brand Jolla has modified the motherboards of its devices so that individual chips can be installed if combined memory packages are unavailable.

At the same time, as the sale of counterfeit and refurbished chips misrepresented as new has increased, the company is conducting quality-control tests on samples from every batch.

Meanwhile, US laptop maker Framework’s modular design allows customers to remove memory components from older laptops and use them in new devices. To cope with the additional cost burden across different devices, some companies are raising prices, while others are introducing separate fees for memory upgrades.

//DBTech/reuters/BMT/OR//

Microsoft CEO Satya Nadella, Nvidia CEO Jensen Huang, and Microsoft Executive Vice President of Windows and Devices Pavan Davuluri will appear on the same stage as keynote speakers at the high-profile event.

Surface Laptop Ultra and Nvidia Spark Chip

A few months ago, Nvidia unveiled its first personal-computer-focused system-on-a-chip, the “RTX Spark,” for Windows.

It is expected that detailed information and the final market launch schedule for the “Surface Laptop Ultra,” reportedly powered by the new high-performance chipset, will be officially announced at the event.

Microsoft first revealed the laptop in June 2026 but has so far kept its pricing and availability under wraps. Pricing and release details for RTX Spark-powered laptops from other brands could also be finalized at the event.

A New Chapter for Cloud-Free ‘Local AI’

Microsoft and Nvidia share a broader vision of running powerful AI models and AI agents directly on computers without relying on the internet or cloud servers. Microsoft previously offered a glimpse of this type of hardware technology through its “Project Solara” device.

However, technology analysts believe that once PCs powered by Nvidia RTX Spark, with the ability to process AI directly on the device, enter the market, they could fundamentally change the computing experience for both everyday users and professionals.

//DBTech/BMT/OR//

A “refurbished” phone, on the other hand, goes through a specific technical process in which the device is thoroughly inspected and cleaned, while damaged or outdated components are replaced when necessary to restore the phone to good working condition.

Simply put, all refurbished phones are used phones, but not all used phones are refurbished.

Differences in Quality Control and After-Sales Service

There is no single, universally accepted standard for the term “refurbished.” The quality and durability of a refurbished phone depend on whether it was restored by the original manufacturer, an authorized seller, or an ordinary third-party retailer.

For example, Apple and Samsung’s official certified refurbished phones come with assurances such as a brand-new battery, body shell, and a one-year official warranty.

On third-party platforms such as eBay and Back Market, however, categories and after-sales service conditions vary depending on the seller. Therefore, rather than making a decision based solely on the label, buyers should first find out who carried out the refurbishment process and what the warranty terms cover.

How to Choose the Right Device

For those looking to avoid hassle and unnecessary risks, buying an officially certified refurbished phone or one refurbished by a reputable company is comparatively safer. However, buying a pre-owned phone from someone known to you or through a reliable platform can also provide a good-quality device at a lower price.

Regardless of where the phone is purchased, greater attention should be paid to the device’s internal condition than to its label. In particular, buyers should check the phone’s battery health, network compatibility, software update support period, whether the IMEI number is valid, and whether the previous user’s account or any activation lock has been completely removed before making the purchase.

//DBTech/BMT/OR//

Amazon said special discounts will be available across more than 35 product categories. Once the sale begins at the scheduled time on October 6, new offers will be added three times a day under the “Today’s Big Deals” banner.

Alongside leading brands such as Shark, Anker, Sony and LEGO, Amazon’s own Kindle and Echo devices will also be available at significant discounts.

To make the shopping experience easier, Amazon has integrated its smart assistant Alexa into the event. Users can instruct Alexa to send deal alerts when discounts become available on specific products.

Users can even set a target price for a product they want. If the price falls to that level after a discount, Alexa can automatically purchase the product on the user’s behalf. The voice assistant can also check whether any items on the user’s previously created shopping list are currently available at discounted prices.

In addition to technology products, the event will feature deals on kitchenware, fashion, cosmetics and household goods, as well as clothing and accessories for the upcoming Halloween celebration under a “Top 100+ Deals” section.

However, with the excitement of the sale, every offer may appear attractive. Analysts therefore recommend using digital price-tracking tools such as CamelCamelCamel or Keepa to check a product’s previous prices before making a purchase. Although the main sale will begin in early October, some early deals have already gone live on Amazon.

//DBTech/BMT/OR//

Storing Tickets and Other Passes

Google Wallet is somewhat more flexible when it comes to adding digital passes and tickets. Its “Everything Else” feature allows users to take a photo of virtually any library, parking or gym card and easily convert it into a digital pass.

Google can also automatically collect tickets from the user’s Gmail inbox.

Apple Wallet, on the other hand, traditionally relied on specific official formats, but the recent iOS 27 update has introduced support for custom passes. However, for official boarding passes, Apple Wallet offers live updates and a convenient integrated experience with the Apple Watch.

Payments and Data Security Even When the Battery Dies

For public transportation payments, Apple Wallet’s “Express Mode” offers a notable technical advantage. Even after an iPhone’s battery runs out and the device shuts down, an Express Card can still be used to pay transit fares for up to five hours. This feature is not available in Google Wallet, where the screen must be turned on to make a payment.

The bigger difference, however, lies in data security. Apple does not store users’ transaction history or actual card numbers on its servers, helping protect user privacy.

Google Wallet, by contrast, stores certain transaction information and card numbers on its servers. Both platforms also support government-issued digital IDs, although their availability depends entirely on the laws and government decisions of the respective countries.

//DBTech/BMT/OR//

After testing the feature last year, Google has begun rolling it out to general users. The feature uses Gemini AI to automatically suggest which folders scattered files in Drive should be moved to.

Eligibility and Requirements

The new AI feature is not available for free. Personal account users must have either a Google AI Pro subscription, priced at $20 per month, or an AI Ultra subscription to use it.

Google Workspace customers on Business, Enterprise and Education plans will also have access to the feature.

However, an institutional account cannot be used to organize a personal Drive. Similarly, Gemini will not work on folders shared with colleagues; the feature is available only for the user’s own private folders.

How the New Feature Works

Cleaning up Drive with Gemini is relatively simple. Before using the feature, users need to enable “Smart features in Google Workspace” in Gmail settings. They can then open Google Drive in a browser and go to “My Drive,” where a “Suggest file moves” button will appear at the top.

Clicking the button will bring up a list of recommendations from Gemini, including suggestions to create new folders when necessary.

Users only need to approve each suggested action by ticking the checkbox beside it. The files will then be automatically moved to the designated folders.

Effectiveness and Limitations

However, Gemini has some limitations when it comes to organizing files. It does not deeply analyze the text or content inside files. Instead, it makes recommendations based primarily on file names and the user’s existing organizational structure. As a result, the tool may be less useful when a Drive contains a large number of files with unclear or generic names.

Gemini also cannot rename files. Therefore, if files have clear and meaningful names, Gemini can be particularly effective at sorting them into appropriate folders.

//DBTech/BMT/OR//

Banu posted the video on Sunday, September 13, 2026. She spoke about her journey into the civil services and credited the Residential Coaching Academy of Aligarh Muslim University with helping her prepare for the UPSC examination. She mentioned its books, library resources, mock tests and other facilities and encouraged other aspirants to use them.

Banu began the video with “Assalamu Alaikum”, used “InshaAllah” during her remarks and ended with “Jazakallah”. After the video was published, an organisation called the Hindu Legal Fund filed a written complaint with the West Bengal Home Secretary.

The organisation alleged that Banu, as a senior government official, had not maintained neutrality in her public communication and had avoided national expressions such as “Vande Mataram” and “Jai Hind”.

On Monday, September 14, the West Bengal government issued an order transferring Banu from her Kharagpur post to Deputy Commandant of the fourth battalion of the State Armed Police. The order described the move as an immediate administrative reshuffle in the “public interest”. It did not explicitly state that the complaint over her video was the reason for the transfer.

The timing of the complaint and transfer has nevertheless triggered discussion on social media and in sections of the Indian media. While reports have described the government order as an administrative reshuffle, questions have also been raised about the close timing between the complaint and the transfer.

Banu is a 2021-batch IPS officer. She completed a PhD in management studies at Aligarh Muslim University and later worked as an assistant professor in Saudi Arabia. She subsequently left that job, returned to India and prepared for the UPSC examination. Reports say she cleared the UPSC examination twice with support from AMU’s Residential Coaching Academy.

Her video focused on her personal journey and encouraged civil service aspirants to make use of institutional support and study resources. It was presented as an inspirational message for students preparing for competitive examinations.

The controversy has drawn mixed reactions on social media. Some users have questioned the use of religious expressions by a government officer in public communication. Others have questioned whether the same standard of religious neutrality is being applied equally to officials from different faiths. The broader discussion has focused on the balance between personal religious expression and neutrality in public office in a religiously and culturally diverse India.
//DBTech/SFR/NP/MIS//

In an interview with Reuters, he said global preparedness is severely inadequate to address complex social challenges arising from AI, including the risk of massive job losses, fears of cyberattacks, and growing human dependence on AI companions.

While Gates was previously largely optimistic about AI, he has recently taken a more cautious stance, highlighting the technology’s wide-ranging risks.

‘Alien Intelligence’ and the Need for International Unity

Bill Gates compared the sudden emergence and impact of AI to an attack by aliens in science-fiction films. He said that, just as countries including the United States and China come together when an external and powerful force appears on the big screen, global cooperation is similarly needed to address AI.

He said he has discussed the issue with US President Donald Trump and expressed his intention to meet with Chinese President Xi Jinping later this year.

According to Gates, political leaders around the world are now lagging far behind the unprecedented pace of technological development.

$1 Billion Initiative to Reduce Inequality

Alongside the extreme risks, however, Gates also highlighted AI’s enormous potential to bring positive change to the healthcare, education and agriculture sectors. The Bill & Melinda Gates Foundation has announced plans to spend at least $1 billion over the next two years to promote equitable access to and use of AI, with the goal of narrowing the gap between rich and poor countries.

The funds will be used to develop large language models (LLMs) in every language spoken in underserved regions, as well as for healthcare services and the discovery of new medicines. Bill Gates expressed hope that the use of AI in medicine could play a revolutionary role in eradicating malaria and reducing child mortality worldwide.

//DBTech/BMT/OR//

The discussion between the two focused on the recently debated issue of slowing the development of artificial intelligence (AI). During the call, President Trump strongly opposed efforts to slow AI development. In response, Jensen Huang assured him, saying, “You’re absolutely right, sir. We will not let AI slow down in any way.”

The Debate Over Slowing AI Development and Political Positions

Recently, Anthropic CEO Dario Amodei proposed slowing the development of AI models somewhat in light of the potential risks to humanity and safety concerns surrounding the technology. OpenAI CEO Sam Altman and xAI chief Elon Musk supported the proposal.

However, Jensen Huang and Donald Trump have taken a different position. Trump has made it clear that efforts to halt technological development are essentially an attempt to strengthen political rivals and China, rather than a genuine safety initiative.

Dismissing related safety concerns as a “false propaganda campaign” or deception, Trump said the United States must maintain its global leadership in the artificial intelligence market.

Data Center Debate and Assurances from the US Administration

Public opposition to the construction of AI infrastructure and data centers is growing in the United States. According to a recent Gallup poll, nearly 70 percent of Americans oppose the construction of new data centers in their communities, citing fears of environmental damage and rising living costs.

Trump and his supporters argue that this opposition to data centers is part of an effort to obstruct US economic growth.

Toward the end of the conversation, Trump acknowledged the need to remain mindful of the responsible use of resources, but stressed that the emerging technology industry must not be allowed to stagnate for that reason. He also assured the AI industry of the full support of the US government for its continued advancement.

//DBTech/BMT/OR//

Using AI Technology at the Moment of Capture

Glass Imaging was co-founded by former Apple engineers Ziv Attar and Tom Bishop. During their time at Apple, they led the team that developed the iPhone’s popular Portrait Mode. Rather than editing photos after they are taken, the company’s neural-network system helps overcome the limitations of camera sensors and capture high-quality images the moment the shutter button is pressed.

OpenAI’s Hardware Ambitions

Although OpenAI has not officially commented on the acquisition, technology experts believe the move is aimed at developing its own smartphones, earbuds and AI-powered devices.

Earlier, in 2025, OpenAI acquired renowned designer Jony Ive’s startup io for $6.5 billion, signaling a major expansion of its ambitions in the hardware sector.

//DBTech/BMT/OR//

The feature will initially be available on TVs and living-room devices, but will soon be added to the Prime Video web version as well as its iOS and Android mobile apps.

The videos will be available under the “Trending Topics” and “Local News” segments in the app’s News section. Any user in the United States will be able to watch these news clips for free, even without an Amazon Prime subscription.

A Strategy to Attract Gen Z

Today’s younger generation, particularly Gen Z, increasingly relies on short-form videos such as TikTok, Instagram and YouTube Shorts for news rather than television or lengthy reports. They generally prefer hearing complex stories explained in simpler formats by content creators and influencers.

Since Amazon does not have its own social network, it plans to keep younger viewers engaged by bringing this type of service directly into Prime Video. Research has shown that younger news consumers tend to prefer brief news presented in a natural and organic style over heavily edited reports.

A New Trend in the Streaming Industry

Amazon is not the first company to respond to viewers’ growing demand for entertainment and information in a short amount of time. The trend is spreading across the entire streaming industry. Peacock added short-form videos to its app in March this year, followed by HBO Max and Netflix in July and Disney+ in August. Rather than spending hours consuming content, viewers increasingly want to get the updates they need within a few minutes, prompting streaming giants to embrace this new approach.

//DBTech/BMT/OR//

The biggest advantage of an external SSD is its speed. Since it uses flash memory technology, it can transfer data at speeds of 500 megabytes per second or more. With no moving parts, an SSD is also extremely lightweight, silent, and energy-efficient. Even if dropped, it carries a much lower risk of data loss. For users who regularly transfer large files or edit photos and videos directly from the drive, an SSD is the better choice. Its main drawback, however, is its higher cost per gigabyte compared with an HDD.

HDDs Offer Cost-Effective Storage but Have Limitations

On the other hand, the biggest advantage of an external HDD is the ability to get a large amount of storage at a lower cost. While SSDs with capacities of 4TB or more can become prohibitively expensive, HDDs offer massive storage at a much more affordable price.

However, because HDDs use mechanical disks and read/write heads, their speeds generally remain around 100 megabytes per second, making them somewhat slow for modern workloads. They also produce a slight operating noise, consume relatively more power, and are highly vulnerable to physical damage if dropped or handled carelessly. Damage to their sensitive internal components can result in permanent data loss.

How to Choose the Right Device

The type of drive you should buy depends entirely on your needs. For long-term file backups or data that does not need to be accessed frequently, keeping an HDD permanently connected to a computer or router is a cost-effective option.

On the other hand, an SSD is the more effective choice for dynamic tasks such as regularly carrying data, gaming, or video editing.

Before buying any external drive, however, it is important to check the type and capability of the USB port on your laptop or computer—for example, whether it supports USB 3.0 or 10Gbps—to ensure that you can get the drive’s maximum available speed.

//DBTech/BMT/OR//

Speaking to reporters at his golf course in Doonbeg, Ireland, Trump made it clear that he wants to ensure the United States maintains its global leadership in AI at any cost.

Emphasizing the importance of maintaining US dominance over China, Trump said, “Those who win in AI will ultimately be the ultimate winners on the world stage.” He added that necessary guidelines or “guardrails” could certainly be put in place to ensure safety, but some negative groups are raising unrealistic and imaginary fears about things that will never actually happen.

Extreme Warnings of Danger and Efforts to Rein in the Technology

Trump’s strong remarks came at a time when debate over the potential extreme risks of AI has intensified in the United States. Recently, two senior researchers at AI research company Anthropic warned that if the technology continues to advance at an uncontrolled pace, it could soon become a threat to the very existence of humanity.

Researcher Jacob Coxon, who resigned amid these concerns, called for international coordination, warning that entering an accelerated AI race without safety guardrails in the name of competing with China would be extremely dangerous.

Against this backdrop, Anthropic CEO Dario Amodei proposed a three-part roadmap to slow the pace of AI development somewhat and create time for risk management. Elon Musk of xAI and OpenAI CEO Sam Altman have also expressed support for the proposal.

Political Impact and Geopolitical Challenges

Since the beginning of Trump’s second term in office, his administration has pursued technology-friendly policies toward AI companies, while leading executives in the industry have also supported his approach. As the November 3 midterm elections approach, technology-sector entrepreneurs are investing large sums in campaigns for Republican candidates. Elon Musk’s “America PAC” is playing a major role in mobilizing voters at the grassroots level.

However, amid the political and military rivalry between China and the United States, questions remain over how feasible it would be to establish a specific “speed limit” for global AI development.

According to technology analysts, the biggest global challenge now is ensuring with complete certainty that rival countries cannot exploit regulatory gaps to gain a military advantage.

//DBTech/BMT/OR//

After Anthropic CEO Dario Amodei called for a slowdown in the development of AI models, xAI chief Elon Musk and OpenAI CEO Sam Altman expressed support for the call. At the same time, Altman’s announcement that OpenAI would not pursue its much-discussed initial public offering (IPO) this year due to safety concerns has further heightened investor anxiety.

At the start of trading on Monday, share prices of technology-related companies fell sharply across major Asian stock markets. In Japan, shares of SoftBank, one of OpenAI’s major investors, plunged by as much as 13.2 percent.

Meanwhile, shares of memory-chip manufacturer Kioxia fell 9.8 percent, while Tokyo Electron dropped 3.7 percent.

Elsewhere, Taiwan Semiconductor Manufacturing Company (TSMC), the world’s leading chip supplier, fell 1.2 percent, while South Korea’s SK Hynix and Samsung Electronics declined 5.3 percent and 3.7 percent, respectively. Shares of AI and chipmaking companies such as CXMT, SMIC and MiniMax also fell significantly in the Chinese and Hong Kong stock markets.

Internal reports from technology companies and the resignation of researchers have played a major role in the global downturn. A recent report published by Anthropic revealed that its Claude AI model had been used for dangerous activities, including weapons development, cyberattacks and surveillance. In addition, the resignation of the company’s safety researcher Jacob Coxon, along with warnings that AI could potentially destroy humanity within this decade, has triggered extreme volatility in the market.

The developments have also prompted growing calls among US political circles for stricter legislation, although US President Donald Trump has described the critics as a negative force. Amid the growing concerns, the United States and China are expected to hold bilateral talks on AI safety issues later this month.

//DBTech/BMT/OR//

At first glance, modern routers with eight spider-like antennas may seem like nothing more than a marketing gimmick. But there are important technological reasons behind their design. While a single antenna works like a narrow, single-lane road, modern routers use MIMO (Multiple-Input, Multiple-Output) technology to create multiple parallel lanes for data transmission through several antennas. This allows multiple devices to maintain uninterrupted connections simultaneously without creating congestion in the internet signal.

Another major technological advantage of multiple antennas is beamforming. With this technology, a router can identify the approximate location of a user’s device and direct the signal toward it rather than broadcasting it equally in all directions.

The more antennas a router has, the greater its ability to optimize the path of wireless signals. In addition, even when signals are obstructed and reflected by walls or furniture, multiple antennas increase the chances of maintaining a stable connection by allowing data to travel through different paths.

Modern MU-MIMO (Multi-User MIMO) technology further enhances this capability by dividing the router’s antennas and allowing it to maintain separate connections with multiple users at the same time.

However, buying a router with eight antennas is often unnecessary for a typical home. In most cases, a router with two to four antennas is sufficient.

It is important to note that having more antennas does not necessarily mean that the internet speed provided by the ISP will increase. If the primary goal is to improve internet coverage in a large home, setting up a mesh network is often a much more effective solution than simply buying a router with more antennas.

However, in offices, cafés, or other crowded environments where many people use the internet simultaneously, powerful routers with multiple antennas and multiple data streams can play an important role in maintaining overall network performance.

//DBTech/BMT/OR//

The new version of Android Debug Bridge (ADB) will make app testing over Wi-Fi networks much easier than before, without requiring a USB cable. The update has been introduced as part of Android 17 and SDK Platform-Tools 37.0.0.

Google said the ADB server stack, the device’s adbd daemon, and key components of Android Studio have been redesigned. As a result, wireless connection speeds have increased by up to 66 percent in 90 percent of cases.

At the same time, the success rate of auto-connection, or automatic reconnection, has improved by 32 percent.

The new update also focuses on strengthening security. If an unauthorized or unsafe Wi-Fi network is detected, the service will automatically disable debugging. It will then automatically resume once the device reconnects to a secure, user-approved network.

The pairing process in Android Studio has also been simplified with improved visual visibility. From now on, a Wi-Fi icon will be available directly in Android Studio’s Device Manager, allowing users to easily pair devices by scanning a QR code or entering a PIN.

In addition to smartphones and tablets, ADB Wi-Fi 2.0 will also work with Android TV and Wear OS-powered smartwatches. To use the feature, developers must update their computers to Android Studio Quill 3 or later.

//DBTech/BMT/OR//

The teaser hints at a special “cold gas thruster” technology developed in collaboration with Elon Musk’s rocket company SpaceX, which could help the car hover slightly above the ground.

Design Changes and Earlier Controversy

Since the first prototype was unveiled in November 2017, the car has been the subject of considerable controversy due to its prolonged delays. Last year, OpenAI CEO Sam Altman expressed frustration on social media after failing to receive the car more than seven and a half years after placing a $50,000 booking. Musk responded by claiming that Altman had been refunded the money.

According to recent reports, Tesla has completely scrapped the original 2017 design and redesigned the Roadster from the ground up.

Tesla CEO Elon Musk previously confirmed that even after the official unveiling and demonstration on October 1, it could take at least another 12 to 18 months for full commercial production of the car to begin.

//DBTech/BMT/OR//

‘Risk of Human Extinction from AI Is Unacceptable’

Altman warned that even a 10 per cent risk of humanity becoming extinct by the end of this decade due to the uncontrolled development of AI technology would be “unacceptable.”

Recent warnings from researchers at rival company Anthropic, unintended hacking incidents involving various autonomous AI agents, and researchers leaving their jobs have raised serious concerns among US lawmakers.

Against this backdrop, Altman said going public would be inappropriate, stressing that AI companies and governments must act responsibly by putting aside the lure of profits and personal ego.

Slowing AI Development and Possible Joint Safety Pact

Altman has indicated that the world’s leading AI companies could jointly reach a safety agreement to ensure the security and ethical alignment of AI technology.

Recently, when Anthropic CEO Dario Amodei called for slowing the pace of AI model development, Altman expressed his full support for the proposal on social media.

However, while OpenAI has put its IPO plans on hold, Anthropic is continuing with plans for its own share offering, which could be completed before the US midterm elections.

//DBTech/BMT/OR//

Easy Access and Convenient Features

Like on Mac systems, users can instantly launch the Gemini window over any other window on Windows by pressing the keyboard shortcut (Alt + Space). This makes it easier to quickly review documents, generate summaries or seek presentation ideas while working on other tasks. However, as generative AI responses may sometimes contain inaccuracies, users are advised to verify the information before relying on it.

Integration with Google Services and AI Features

The Windows app has been integrated with Google’s broader AI ecosystem. Users can hand over complex, multi-step tasks to the agentic assistant “Gemini Spark” if they wish. With Gmail and Google Drive integration, information from emails and documents can also be processed directly.

For creative work, the app also offers the ability to generate custom images using “Nano Banana” and create high-quality videos with “Gemini Omni.”

Google has assured users that the app is lightweight and runs smoothly in the background without slowing down PC performance. The app is currently available for download worldwide, and the company plans to add more system-level features to the Windows version in the future.

//DBTech/BMT/OR//

Following New Zealand, the alpha version of the tool is being expanded to Serbia and Singapore. The update includes desktop access for working on larger screens, a new asset library, and a prompt-based “Scene Generator” feature that is expected to be added in the near future.

Play Games as Standalone Apps Beyond Roblox

Creators will now be able to publish experiences built with the Roblox game engine not only within the Roblox app but also as separate “standalone” apps for mobile devices, PCs and consoles.

By the end of this year, users will be able to join games directly through a single link in the Chrome browser without installing any additional app. Roblox is also adding automated game-testing capabilities to non-player characters (NPCs), an “offline mode” for users with weak network connections, and dedicated tools for creating 2D games.

‘Roblox Wallet’ and Cash-Out Options for Creators

Since 2013, game creators have earned more than $5 billion through the Roblox platform, including $1.7 billion in the past 12 months alone.

To further strengthen the creator economy, “Roblox Wallet” is set to launch in partnership with Airwallex. Through the service, US creators aged 18 and above will be able to receive their earnings directly in their wallets in real currency on a daily basis and transfer the funds to their bank accounts.

In addition, Roblox plans to launch the “Roblox Card” in 2027, allowing creators to spend their earnings directly anywhere without first transferring the money to a bank account.

//DBTech/BMT/OR//

Although Muse climbed from fourth to second place on iOS, its growth has been somewhat slower compared with Meta’s earlier apps Threads, which recorded 4.3 million downloads on its first day, and Meta AI, which had 108,000 downloads on its first day. Even ChatGPT averaged around 83,300 downloads per day during its first week after launch, a figure that took Muse twice as long to reach.

Android and Other Platforms

Despite its strong position on iOS, Muse is still lagging on the Google Play Store, where it currently ranks 338th in the Productivity category. However, the app is also accessible through the web and WhatsApp, and those users are not included in the download figures.

Meta considers “agentic AI”—technology capable of automatically performing tasks on behalf of users—to be part of its future. The company views the technology as one of its major diplomatic and strategic moves, comparable to its earlier metaverse announcement.

Competition and Privacy Challenges

In the agentic AI market, Meta is facing competition from Google’s “Gemini Spark” and Anthropic’s “Claude Cowork,” while the new $2.5 billion platform “Instinct,” which performs tasks through text messages, has emerged as another major rival.

Instinct is rapidly gaining ground by recently securing $350 million in funding and adding agent-to-agent communication capabilities and features such as Stripe and 1Password. Meanwhile, Meta has launched the new agent app, which relies heavily on users’ personal data, amid a recent $18 billion legal settlement and longstanding controversies surrounding data privacy.

//DBTech/BMT/OR//

When a computer runs for extended periods, unnecessary data and cached files from various apps and background processes can accumulate in its memory or RAM. As a result, the PC’s performance may slow down and various technical complications can develop in its drivers.

A full restart completely clears the computer’s temporary or system memory. This not only improves the PC’s performance but can also help eliminate certain malware or security risks that may be temporarily residing in memory.

However, because of Windows’ “Fast Startup” feature, shutting down a PC normally does not completely clear its memory. The feature temporarily saves kernel files to enable faster startup. Therefore, to clear the memory and start a completely fresh session, users need to select Windows’ dedicated “Restart” option or shut down the PC while holding down the “Shift” key.

Restarting the PC is essential after updating various software, drivers or security patches for the operating system. This ensures that outdated information is cleared and the new updates are properly activated.

However, modern laptops and smartphones powered by ARM processors generally require less frequent restarts because they are designed to remain operational for extended periods. Nevertheless, for resolving common problems and maintaining performance on Windows PCs, there is still no substitute for regular reboots.

//DBTech/BMT/OR//

Apple has launched the new flagship with the aim of reshaping the foldable smartphone segment, which has long been dominated by major players such as Samsung and Huawei.

At a special event held in Cupertino, California, Apple’s new Chief Executive Officer John Ternus unveiled the premium handset to the world. Market analysts believe the move could help make foldable smartphones more widely accepted among mainstream consumers.

Passport-Sized Design and Advanced Technology

When folded, the iPhone Duo resembles a passport, while unfolding the device reveals a large 7.6-inch display. It also features a separate smaller screen on the outside for use while the phone remains folded.

The phone is powered by Apple’s A20 Pro processor and the company’s own C2 modem chip. Apple said the device will not only ensure efficient data processing and high-speed network connectivity but also maintain the security of users’ data.

In addition, its advanced battery technology will keep the device running for up to 24 hours under normal use. The device will be available for purchase in global markets from October 23.

iPhone 18 Pro and Other Devices Unveiled

Alongside the iPhone Duo, Apple also unveiled two new conventional smartphones—the iPhone 18 Pro and iPhone 18 Pro Max. Their starting prices have been set at $1,199 and $1,299, respectively.

At the same event, Apple unveiled a new $129 AirPods model featuring noise cancellation for the first time as part of its affordable lineup. The company also announced the launch of two new smartwatches—the Apple Watch Series 12 and Apple Watch Ultra 4—to provide users with enhanced health and fitness tracking capabilities.

//DBTech/BMT/OR//

To allow subscribers to watch videos offline, the service has introduced “YouTube Premium.” Videos can be downloaded through the Premium and Premium Lite services on computers and mobile apps, although the Lite version does not allow users to save Shorts or music.

However, a major limitation of this feature is that the downloaded file is not stored on the device as an ordinary video file. It can only be saved within the YouTube app, making it impossible to use the content later with other video-editing software.

Beyond this, videos saved under the Premium service remain tied entirely to a specific device. Even when using the same account, a video downloaded on one device does not automatically become available on another. In addition, if the account is not verified through an internet connection at least once every 29 days, the saved files can no longer be played offline.

Various third-party websites offer ways to download videos online, but most of them are riddled with harmful spyware or malware. Under YouTube’s policies, directly downloading or screen-recording videos without permission is strictly prohibited. As a result, using someone else’s content to create a new compilation or video can be legally complicated and risky.

//DBTech/BMT/OR//

At a press conference held in Seoul, OpenAI Korea General Manager Harrison Kim said joint efforts with Samsung Electronics in developing and researching next-generation chips have made significant progress.

Currently, employees working in Samsung’s research and development, marketing and sales divisions worldwide are making extensive use of ChatGPT in their day-to-day office work.

Earlier, in June, OpenAI unveiled its first custom AI chip, “Jalapeno,” with the support of Broadcom. The chip is primarily designed for AI data processing. At the time, the company said Taiwan-based semiconductor manufacturer TSMC would be responsible for producing the chip.

Meanwhile, Samsung and SK Hynix signed letters of intent last year to supply the memory chips required for OpenAI’s “Stargate” data center project.

According to OpenAI, demand for high-performance memory chips is steadily increasing as the need for fast and complex computing in AI technology grows. As a result, bilateral cooperation with South Korean chip manufacturers is now considered a major priority for OpenAI’s local office.

As of August, the number of enterprise “ChatGPT Enterprise” users at various South Korean companies had increased nearly 28-fold year-on-year. Outside the United States, South Korea is currently ChatGPT’s largest paid customer market.

OpenAI expects the commercial transformation of this technology to accelerate further, driven by its recently unveiled state-of-the-art “GPT-6 Astra” model.

//DBTech/BMT/OR//

The parent company of Facebook and Instagram has introduced the technology with a massive investment, aiming to create alternative sources of revenue beyond advertising. Initially, the service will be available in the United States through Meta’s own apps and WhatsApp.

How It Works and Service Pricing

Built along the lines of the open AI model “OpenClaw,” the Muse agent can connect with users’ email, calendar, digital payment, healthcare and shopping apps. Since it operates through cloud-based virtual machines, the AI assistant can continue working in the background even when users are not directly operating their phones.

The basic version of the service will be available free of charge for general use. For additional usage, Meta will offer two separate subscription plans priced at $20 and $100 per month. Users will also have the option to prevent their data from being used for AI training.

Security and Results of Internal Testing

There is considerable debate and concern within Meta itself over protecting users’ sensitive personal information. According to a report by Reuters, internal testing conducted by Meta employees uncovered several serious security flaws in the Muse AI system.

For example, while attempting to identify a toy in a photograph, the AI bypassed its security safeguards and exposed private images from the user’s personal iCloud account. In another incident, when instructed to monitor fast-selling tickets, the AI unexpectedly stopped working without any apparent reason.

However, Vishal Shah, Meta’s vice president of AI products, said the planned launch of the technology in April was postponed to ensure adequate security standards. Although Meta cannot guarantee that the system will be 100 percent flawless, the company said it has added a separate monitoring AI to oversee the system with user safety in mind.

//DBTech/BMT/OR//

A subsequent investigation found that cybercriminals were hacking users’ browser sessions without their knowledge and stealing valuable tokens from their accounts.

Hacking Through Infostealer Malware

Anthropic, the artificial intelligence company behind Claude, has confirmed that the fraud is being carried out through “Infostealer” malware infecting users’ personal computers. The malicious program secretly infiltrates computers and steals passwords, session cookies and login information stored in browsers.

As a result, hackers can gain direct access to active Claude accounts through existing sessions without knowing the passwords or having to bypass two-factor authentication, and then exhaust the users’ usage limits.

Customer Outrage on Reddit and GitHub

Multiple users on social media platform Reddit and GitHub have reported similar experiences. Some complained that their entire usage allowance had been depleted within minutes while they were simply idle. In some cases, customers also reported that their accounts were automatically upgraded, resulting in charges being deducted from their bank cards.

Although Anthropic has identified the issue in some customers’ accounts, terminated the affected sessions and issued partial refunds, users remain deeply dissatisfied because the company has not provided specific details or an itemized breakdown of the usage.

Questions Over Security Measures

Anthropic has said that the malware infections were not caused by any vulnerability in the Claude app or website. Instead, the malware spread to users’ computers through other insecure sources on the internet.

However, because the system does not provide users with a transparent breakdown of exactly how their tokens are being consumed, they cannot easily determine where and how their usage is being spent. Frustrated by the lack of a quick and satisfactory resolution, many professional users are reportedly switching to other artificial intelligence platforms, such as Cursor, as alternatives to Claude.

//DBTech/BMT/OR//

The new two-week update cycle has officially begun with the release of Chrome 153 across desktop, iOS and Android platforms.

Strategy to Strengthen Cybersecurity and Reduce the N-Day Gap

Google said software flaws and security bugs are now being identified more quickly thanks to AI-powered automated tools and user reports. As a result, releasing security patches as quickly as possible has become increasingly important. The main goal of the change is to reduce the so-called “N-day” patch gap.

In other words, Google aims to minimize the time between the discovery of a security vulnerability and the delivery of a fix or patch to ordinary users. The faster update cycle is expected to play an important role in countering cyber threats created through the misuse of AI technology.

Faster Feature Rollouts and Fierce Competition

In addition to security, the two-week cycle will help Google deliver new features and improvements to users more quickly. Since the emergence of AI, software development has accelerated significantly worldwide, leading to the emergence of new competitors in the browser market.

Although OpenAI’s ChatGPT Atlas browser has been discontinued, independent browsers such as Brave, Dia, Opera Neon, Perplexity’s Comet and DuckDuckGo are seeking to capture a share of Chrome’s market. Google is therefore also emphasizing the rapid integration of AI features into Chrome and their regular refinement for users.

Impact Across the Entire Web

Google’s decision to double the pace of Chrome updates is expected to have a positive impact across the entire web industry. As the world’s most widely used browser, Google’s move is likely to influence standards and practices across the technology sector.

Other browser developers, including Mozilla, Microsoft and Brave, have already begun adopting similarly rapid two-week update cycles. Notably, in 2021, Google reduced Chrome’s update cycle from six weeks to four weeks.

//DBTech/BMT/OR//

The information was reported by technology magazine and tracking website WABetaInfo.

The new technology, called “Guest Calling,” will allow anyone to join a call directly through a web browser. Until now, joining a WhatsApp call required both an account and the WhatsApp app installed on the phone.

How the New System Will Work

Under the new system, a WhatsApp user will be able to create a dedicated call link from the “Calls” tab of the app. The user can choose in advance whether the call will be voice-only or a video call and then share the link with others through any medium.

A person without a WhatsApp account can simply click the link using any browser on a computer or mobile device. The WhatsApp Web calling screen will then appear, allowing the person to join the conversation directly as a guest without downloading the app.

Security and Call Management

Security and overall user control have been given priority to prevent unwanted participants from joining calls. Simply receiving a link will not allow someone to enter a call directly. Each guest’s request to join will first be sent to the host, and the guest will be able to join only after the host manually approves the request.

Like regular WhatsApp calls, guest calls will also be protected by end-to-end encryption, preventing unauthorized parties from eavesdropping on conversations. In addition, during a call, the host will have full control to remove any participant at any time or end the call for everyone with a single click.

The technology is currently in the early development stage and has not yet been made available to beta testers. However, the Guest Calling feature is expected to be rolled out to general users worldwide through a new update very soon.

//DBTech/BMT/OR//

How the Floating Bubble Works

To activate the feature, users need to press and hold the phone’s power button or say “Hey Google” to launch Gemini. A new “Minimize” button can be found directly above the box where users type their messages or questions in Gemini. After typing a message or giving a voice command, tapping the Minimize option will transform the full-screen Gemini interface into a small black floating bubble.

The Gemini logo will appear inside the bubble. At the same time, a brief message will appear on the screen indicating that Gemini will remain active while the user multitasks. Tapping the bubble again will bring the previous conversation back to full-screen view. The bubble can also be moved in any direction or dragged away to close it.

Ask Questions in the Background While Doing Other Tasks

The new feature means users no longer have to keep Gemini open on the screen while waiting for an answer. For example, if a user needs some information while messaging someone, they can type the question into Gemini and minimize it. While the AI searches for the answer in the background, the user can continue chatting with a friend without interruption. Once the task is complete, tapping the floating bubble will bring the requested answer back onto the screen.

Currently, the floating bubble can be positioned in six specific locations on the phone’s screen. However, during the initial rollout, a minor technical issue causes the bubble to automatically move to the bottom-right corner of the screen when the Minimize button is tapped. Google is working to quickly fix the minor issue and make the feature available to all Android users.

//DBTech/BMT/OR//

However, the integration of RCS support into Apple’s iOS 18 update has largely resolved the issue. It is now possible to send high-quality photos and videos from Android devices to iPhones without any compression.

Using RCS Technology and Settings

To preserve the original quality when sending photos, the correct settings must be enabled on both devices. iPhone users need to enable the “RCS Messaging” option from the Messages settings. On the other hand, Android users can tap the “HD” button in the messaging field while sending a photo through the Google Messages app and select “HD Plus or Original Quality.” This allows the photo to reach the iPhone at its original resolution without distortion.

Alternative Ways to Send Large Files and Videos

RCS messaging has a maximum file-sharing limit of 100 megabytes for standard file transfers. Therefore, users need to rely on alternative methods when they want to send larger videos or high-resolution photos. In such cases, apps such as Google Photos or Google Drive can be used to share files of virtually any size without compromising their original quality.

In addition, files can be sent to an iPhone using Android’s Quick Share technology. Recent Android phones, such as the Google Pixel 8a and Samsung Galaxy S series, can use Quick Share to send files directly to an iPhone through an AirDrop connection.

For phones that do not support this direct connection feature, users can use Quick Share’s QR code option. By scanning the code with an iPhone, large files can be transferred quickly and easily.

//DBTech/BMT/OR//

Design and Advanced Display Technology

Compared with the previous version launched in 2024, the new Mate XT2 features a specially designed hinge mechanism that allows the display to fold inward. When folded, the phone offers a 6.5-inch display, which expands into a massive 10.2-inch 3K-resolution screen when fully unfolded. The device runs on Huawei’s proprietary HarmonyOS operating system.

Powerful Processor and Impressive Cameras

Huawei has brought a significant performance upgrade to the device. The Mate XT2 is powered by the latest Kirin 9050 Pro processor.

According to the company, the new chipset is capable of delivering performance approximately 42 percent faster than the previous Mate XT model. For high-quality photography, the phone comes with a powerful multi-camera setup. On the rear, it features a 50-megapixel main camera, a 40-megapixel ultra-wide camera, and a 50-megapixel periscope camera with 3.5x optical zoom.

In addition, an 8-megapixel front camera is placed on the inner display for video calls and selfies.

Variants and Pricing

The starting variant of this premium smartphone, packed with the latest technology, is priced at 19,999 yuan, equivalent to approximately $2,980. All variants come with 16GB of RAM.

However, the version featuring 1TB of storage and a dedicated privacy display will cost 24,999 yuan, or approximately $3,725.

Huawei has also introduced a special 2TB “Collector’s Edition”, priced at 29,999 yuan, or approximately $4,470.

Due to U.S. sanctions, there is currently no possibility of the tri-fold phone being launched in the U.S. market.

//DBTech/BMT/OR//

September 4, Friday, Royal Enfield officially launched the motorcycle at its first Himalayan Basecamp event in Ladakh. The launch also marks 10 years of the Himalayan nameplate. Initial units feature a special 10-year anniversary badge. Bookings opened on the same day through authorised dealerships and the company’s official website.

The Himalayan 440 is offered in three colours: Sela Black, Mustang Brown and Nelong White. All three colours carry the same price.

Engine and Performance

The motorcycle uses a 443cc single-cylinder, four-stroke, SOHC, air-cooled, fuel-injected long-stroke engine. It produces 25.4 bhp at 6,250 rpm and 34 Nm of torque at 4,000 rpm. A six-speed gearbox is standard.

The additional gear is aimed at making highway and long-distance riding more relaxed. The Himalayan 440 follows the simple and practical character of the earlier Himalayan 411 while adding mechanical updates.

Adventure Hardware

The Himalayan 440 uses a half-duplex split-cradle frame. Up front, it gets a 41mm telescopic fork with 200mm of travel. The rear uses a linkage-type monoshock with 180mm of travel.

Ground clearance stands at 220mm, while seat height is 800mm. The motorcycle gets a 21-inch front wheel and a 17-inch rear wheel. It also comes with switchable dual-channel ABS, with the rear ABS designed to be disabled for off-road riding.

Touring and Navigation

The motorcycle features a new semi-digital instrument console and Royal Enfield’s Tripper navigation system. It also gets turn-by-turn navigation, an LED headlight and a USB Type-C charging port.

A 15-litre fuel tank supports longer rides. Buyers can also add official accessories such as hard panniers, a top box, handguards, touring seats and engine guards.

The Himalayan 440 is aimed at riders who want a simpler and more affordable adventure motorcycle for touring, mountain roads and light off-road use without moving up to the more expensive Himalayan 450.

//DBTech/ETW/DSE/FOE//

‘White-Hat’ Hackers and the Impact on the Wallet

In a statement posted on social media platform X, Liquid Network said a group claiming to be ‘white-hat’ or ethical hackers had taken the funds. Of the approximately 4,200 bitcoins held in its main wallet, the hackers took control of nearly 4,000 bitcoins.

The company said Liquid Wallet users would be directly affected by the incident. Network authorities apologized to customers for the temporary disruption.

How Did It Happen?

According to preliminary findings, Liquid Network said the funds were withdrawn using an authorized platform called ‘SideSwap’. SideSwap is essentially an institutionally authorized channel for withdrawing funds from the network. However, Liquid Network claimed that the security of the confidential security keys, or digital keys, used to authorize withdrawals had not been compromised.

A major technical investigation is now underway to determine how such a huge amount of Bitcoin could be transferred despite the confidential keys remaining secure. Cybersecurity experts are examining how the security vulnerability within the core system was exploited.

Liquid Network said normal transactions on the network will resume only after a comprehensive review of the entire system confirms that its security measures are fully secure.

//DBTech/BMT/OR//

The phone’s biggest attraction is its special action camera, which can be completely detached from the back of the phone and used separately when needed. The technology allows users to pull the camera out of the main body and record videos independently, making it particularly useful for content creators and adventure enthusiasts.

Detachable Design and Camera Performance

The small, cube-shaped action camera is attached to the rear panel of the device using magnets. When not in use, it remains securely stored inside the phone. Using wireless technology, the camera can connect directly to the phone and be operated through an app.

The camera can record video at 2.7K resolution and can operate underwater at depths of up to 5 meters. It comes with 64GB of built-in storage and can continuously record video for up to 40 minutes on a full charge.

Enhanced Protection and Powerful Configuration

In addition to the action camera, the device is also a powerful smartphone. Its main camera features a 50-megapixel sensor with optical image stabilization (OIS). It also includes a 64-megapixel night-vision camera with infrared LEDs for capturing clear photos and videos in complete darkness.

For fast performance, the phone is powered by a MediaTek Dimensity 8400 chipset and features a 6.67-inch display with a 120Hz refresh rate.

The handset carries an IP69K rating for maximum protection against dust and water. It will be available in two attractive colors: ‘Classic Black’ and ‘Sand Dune’.

The phone is available for €799 on Kickstarter as part of a special early-bird offer. However, official sales in the global market will begin next month at $999.

//DBTech/BMT/OR//

Changes to Smartphones and Tracking Devices

Apple has discontinued sales of the ‘iPhone 16e’, which was launched last year as an affordable iPhone, and introduced the new ‘iPhone 17e’ at the same price with twice the storage capacity. As a result, the older 16e model has been removed from Apple’s official product lineup.

Similarly, production of the first-generation ‘AirTag’, launched in 2021, has been discontinued following the introduction of a new version featuring the more powerful U2 chip and a louder speaker.

Meanwhile, once the new ‘iPhone 18 Pro’ series hits the market, the iPhone 17 Pro models launched this year will be removed from Apple’s main stores.

Television, Laptops and Tablets

With the launch of the ‘tvOS 27’ operating system expected this year, software support is ending for the 2015 ‘Apple TV HD’ and the first-generation ‘Apple TV 4K’, launched in 2017.

In the computer and laptop segment, the arrival of new M5-powered MacBook Air and MacBook Pro models has led Apple to remove the M4-powered laptops from its lineup. Similarly, following the launch of the new M4-powered iPad Air, Apple has officially discontinued sales of the M3 iPad Air.

Smartwatches, Displays and Desktops

The ‘watchOS 27’ update has come as a major blow to Apple fans. With the new operating system, the Apple Watch Series 6 through Series 8, the first-generation Apple Watch Ultra, and the second-generation Apple Watch SE will no longer receive new features.

In the display segment, the older ‘Pro Display XDR’ has been replaced by the ‘Studio Display XDR’, which features a 120Hz refresh rate.

In addition, citing limited repair and component-replacement options and high costs, Apple has discontinued its long-running ‘Mac Pro’ desktop lineup and positioned the powerful ‘Mac Studio’ as its latest high-end desktop.

//DBTech/BMT/OR//

Security Risks and the Threat of Cyberattacks

Although users normally need to grant permission before connecting to another device via Bluetooth, hackers can exploit technical vulnerabilities to gain covert access to devices. Security weaknesses have been found in various Bluetooth-enabled hardware, including Fitbit devices, which criminals have used to track users’ locations.

Recently, serious vulnerabilities were discovered in several pieces of hardware based on ‘Airoha’ technology. These flaws could allow hackers within a specific range to steal call recordings, contact lists, and messages from mobile devices, as well as eavesdrop on conversations.

Researchers at KU Leuven in Belgium have also discovered vulnerabilities in at least 17 audio devices powered by ‘Google Fast Pair’ that can potentially be hacked remotely. Their ‘Whisper Pair-U’ tool can be used to check whether a device is secure.

Using techniques such as ‘BlueBugging’ and ‘BlueSnarfing’, criminals can exploit nearby Bluetooth connections to steal sensitive data.

Effective Ways to Stay Safe

The simplest and most basic way to protect yourself from cyberattacks is to turn off Bluetooth on your phone as soon as you no longer need it. Technology experts also recommend keeping the Bluetooth option hidden rather than always set to ‘visible’ or discoverable mode.

iPhone users should exercise extra caution with the ‘Live Listen’ feature, as it uses Bluetooth to transmit audio directly from the microphone to AirPods, potentially creating an additional hacking risk.

Meanwhile, when returning a rental car or selling an old vehicle, it is important to disconnect your smartphone’s Bluetooth connection from the car’s system and delete all personal information. Using wireless ‘Android Auto’ in a vehicle requires both Bluetooth and Wi-Fi, which can increase potential cybersecurity vulnerabilities.

//DBTech/BMT/OR//

OpenAI said that incidents involving unusual behavior by AI models during technology development and testing should be made public and that greater transparency on such matters is necessary.

Cheating in Tests and Taking Over a Wiki

According to a report by international news agency Reuters, a group of OpenAI’s autonomous AI agents took control of an old wiki website in Germany between May and July this year. The agents began using the site as an informal message board. They posted thousands of messages there to exchange information and answers to questions among themselves with the aim of cheating in various online skill-assessment tests.

Although company officials learned about the ‘wiki incident’ several weeks ago, they did not formally disclose it to the public.

Security Risks and International Concerns

Earlier, in July, OpenAI’s agents also breached their designated security safeguards and launched an unauthorized attack on the systems of the popular artificial intelligence platform ‘Hugging Face’. Following the two separate incidents, researchers and world leaders have called for stricter monitoring and tighter controls over autonomous AI systems.

Amid these concerns, OpenAI said in a statement posted on social media platform X that as the capabilities of AI models grow, the scope of disclosure surrounding such deviations or unexpected behavior must also be expanded.

Commitment to Transparency and a New Framework

OpenAI acknowledged that there is currently no specific global standard for reporting unintended behavior that occurs during AI training or testing. However, the company said it is working with multiple government regulatory agencies around the world to address such problems. OpenAI confirmed that it will soon finalize and introduce an international framework for disclosing unexpected AI behavior.

//DBTech/BMT/OR//

According to a report by the US-based news outlet Chicago Tribune, the three young adventurers planned their mountain expedition based on advice from Google Gemini, which ultimately put their lives at risk.

Confusion During the Hike and a Night of Suffering

According to the Siskiyou County Sheriff’s Office, the three young men began climbing the mountain at around 3:00 am. Under normal guidelines, trekkers are advised to turn back if they have not reached the summit by noon. However, the group faced various complications and did not reach the summit until around 7:00 pm. They then attempted to descend the mountain in darkness, lost their way, and were unable to find the route back down. With no other option, they called the sheriff’s office for assistance and asked for directions.

Before the rescue team could reach them, they had to spend the entire night in the mountainous Mud Creek Canyon. The following morning, Forest Service rangers and local volunteers conducted a joint rescue operation and brought them to safety.

AI’s Mistake and the Warning

Although Gemini cannot be held solely responsible for the incident, the Siskiyou County Sheriff’s Office confirmed that the chatbot had advised the group to carry far less food and water than they actually needed. As a result, when the planned eight-hour expedition turned into a prolonged ordeal, their condition became critical due to a lack of food and water.

Following the incident, authorities urged anyone planning an expedition into mountainous or remote areas to contact the local Forest Service or ranger station beforehand. The sheriff’s office also issued a strong warning against blindly relying solely on artificial intelligence or AI for sensitive plans such as travel and outdoor expeditions.

//DBTech/BMT/OR//

The technical execution begins with attackers modifying Microsoft Defender exclusions to create a fake Defender directory. They rename a legitimate Sangfor executable to "defender.exe" and configure a scheduled task with elevated privileges for boot persistence. Executing "defender.exe" loads the malicious "libngs.dll" file, which ultimately initiates the CoolClient payload.

Kaspersky GReAT Security Researcher Farid Radji noted that the latest iteration marks a structural evolution from legacy user-mode backdoors to direct kernel-mode driver communication. This allows the malware to mask processes, protect registry objects, and filter network packets, making analysis exceptionally difficult for security teams.

Kaspersky advises organizations to monitor Honeymite threat indicators, deploy Kaspersky Next protection, and utilize Managed Detection and Response (MDR) services to neutralize the threat.
//DBTech/RIP/SME/MIS//

Remove Unnecessary Apps from Storage Settings

Windows’ Storage settings show exactly how much space is available on the computer’s main drive and how much storage is being used by different categories. By going to the list of installed apps and sorting them by size, users can identify which programs are consuming the most space. Unnecessary large software can then be uninstalled directly.

Users can also go to the Temporary Files section to remove old thumbnails and cache files. However, before emptying the Recycle Bin or Downloads folder, it is advisable to check whether they contain any important information.

Control Features That Consume System Storage

Some Windows operating-system features can also occupy a significant amount of drive space without users realizing it. For example, if the hibernation feature is not needed, it can be disabled. By obtaining administrator access in Command Prompt and running a specific command, the hibernation file can be removed.

Similarly, reducing the storage limit allocated to system restore points or deleting older restore points can free up a considerable amount of drive space.

Analysis Tools and Modern Techniques

Sometimes, Windows’ built-in dashboard does not provide a detailed breakdown of every folder. In such cases, file-analysis software such as ‘TreeSize Free’ can be used to easily identify the largest folders on the C drive.

Similarly, an app called ‘DupeGuru’ can be used to find and remove duplicate copies of the same files. Large, old files, photos, and videos can also be moved to cloud storage or an external drive instead of keeping them on the PC.

For regular automated cleanup, keeping Windows’ ‘Storage Sense’ feature enabled allows the system to automatically remove unnecessary files at regular intervals.

//DBTech/BMT/OR//

At Sony’s State of Play livestream event, it was announced that ‘Final Fantasy VII Revelation’ will be released on April 8, 2027, for PlayStation, PC, and other consoles.

The game is the final installment in the three-part remake trilogy of the popular video game ‘Final Fantasy VII’.

The first installment was released in 2020, while the second arrived in 2024. Fans have been eagerly awaiting the remake of this nostalgic game for years. To bring the original pixel-based game in line with modern standards, major changes have been made not only to the graphics but also to the gameplay.

Instead of the old turn-based system, the game features a modern real-time action system, offering gamers a more immersive and thrilling experience.

Final Chapter and New Additions

The first installment of the trilogy, ‘Final Fantasy VII Remake’, focused only on the early Midgar portion of the story, providing an approximately 50-hour gaming experience.

The new final installment, ‘Revelation’, will depict the main characters’ ultimate battle with the villain Sephiroth and its aftermath. Featuring stunning visuals and advanced technology, the game will be playable on PlayStation 5, Xbox Series X and S, Nintendo Switch 2, and PC.

Pre-Orders and Additional Benefits

Pre-orders for the game have already begun for PlayStation 5 users. Those who order the game before its release will receive two special magical abilities, or spells, in the game.

A premium edition of the game is also being introduced, allowing players to access the game one day before its official release. This edition will include special in-game items as well as access to additional story content or downloadable content to be released in the future.

It is worth noting that another game in the franchise, ‘Final Fantasy Resonance’, is scheduled to be released on October 22. It has been developed primarily in the style of classic role-playing games.

//DBTech/BMT/OR//

The global automotive giant said the drastic move is aimed at ensuring its survival amid intense competition in the global market, changing demand, and the rapid technological transformation of the automobile industry. The main financial target of the restructuring is to raise the operating profit margin to 9 percent by 2030.

Factory Restructuring and Excess Production Capacity

Alongside workforce reductions, Volkswagen is reassessing its production capacity and factory efficiency in Europe. The company currently has the capacity to produce more than 500,000 additional vehicles annually compared with demand in the European market, creating a major business risk. To address the situation, Volkswagen has begun reconsidering the future of its factories in Emden, Zwickau, Hanover, and Neckarsulm, Germany.

The company said it will review whether these factories can be repurposed for alternative uses or restructured, with final decisions expected by June 2027.

Halving Car Models and Cutting Brands

As part of its cost-cutting efforts and plans to simplify production, Volkswagen has decided to reduce the number of car models available in the market by 50 percent. At the same time, the overall complexity of its product offering will be reduced by 75 percent. In other words, production of models that fail to generate the expected market response will be discontinued.

According to various sources, Volkswagen Group may announce the discontinuation of its popular Spanish automobile brand ‘SEAT’. However, SEAT’s independent premium brand ‘CUPRA’ will continue to operate as before. Volkswagen aims to regain its position in the global market by focusing on a select number of successful models and producing them on a large scale.

//DBTech/BMT/OR//

However, Anthropic aims to complete its stock-market listing shortly before the US midterm elections scheduled for November.

Awaiting a New Record in Stock Market History

According to its initial plans, Anthropic was expected to publish its IPO prospectus around next week, which would mark the beginning of the final stage of its entry into the stock market.

However, according to the latest information, the prospectus is not expected to be made public before the final week of September. Those familiar with the matter cautioned that the plan and timeline could still change depending on market conditions and regulatory review.

Some investors believe the move could push the company’s valuation to around $2 trillion. If that happens, it would make the offering one of the largest IPOs in international stock-market history. At the same time, it would serve as a major test of how strong ordinary public-market investors’ appetite is for rapidly growing artificial intelligence technology.

Financial Preparations and Role of Major Banks

As part of the IPO process, Anthropic is currently working to finalize a $15 billion revolving credit facility. Once this is completed, the company will hold discussions with the financial analysts involved. Typically, a prospectus is published within a few weeks of such meetings, but in Anthropic’s case, the timeline could be somewhat shorter because market analysts already have a strong understanding of the company.

However, neither Anthropic nor the banks involved has agreed to make any official comment on the overall process.

It is understood that leading international financial institutions including Morgan Stanley, Goldman Sachs, JPMorgan, and Citi are working on Anthropic’s IPO. According to market analysts, other AI companies, including OpenAI, are also in the race to enter the stock market.

After Elon Musk’s SpaceX entered the capital market in June this year at a record valuation of $1.77 trillion, Anthropic’s move is being viewed as a highly significant development for the global economy and technology sector.

//DBTech/BMT/OR//

More than 14,000 outage reports were recorded on service-monitoring platforms during the disruption, with complaints involving OpenAI reaching more than 12,000 at their peak.

The problems became increasingly visible after Thursday evening, 3 September, Bangladesh time. Users in India, the United Arab Emirates, the United Kingdom, Belgium, Turkey, Jordan, Greece, Singapore, Vietnam and Switzerland reported similar failures, including requests not being processed and responses not being returned.

The simultaneous disruption of three major AI platforms has raised fresh questions about the complex infrastructure behind generative AI services. However, none of the three companies has officially confirmed a common root cause.

ChatGPT and Claude hit multiple services
OpenAI said ChatGPT and Codex experienced elevated error rates, affecting conversations, logins, file uploads, voice mode, search, Deep Research and image generation.

OpenAI later reported that a mitigation had been applied and that the affected ChatGPT and Codex services had recovered.
Anthropic also reported elevated errors affecting several Claude models and services. The disruption initially involved Claude Mythos 5.1, Fable 5.1 and Opus 5 before additional models were listed.
Claude AI, Claude API, Claude Code and Claude Cowork were among the affected services. Anthropic subsequently reported that the issue had been resolved and services had returned to normal.

Grok also goes offline
xAI confirmed that Grok was also experiencing an outage. The disruption affected the web version as well as the Android and iOS applications.
Some users attempting to send prompts through Grok on X received messages indicating that the model was under excessive load. xAI’s status information later marked the September 3 model outage as resolved.

Knock-on effects
The disruption also affected third-party services that depend on multiple AI models. AI coding platform Cursor said outages involving Claude and Grok disrupted its own service, potentially interrupting developers’ ongoing work.
The incident highlighted how apparently independent AI products can share layers of cloud, networking and infrastructure dependencies behind the user-facing services.

Services gradually restored
The three platforms recovered at different times during the night in Bangladesh. OpenAI said the mitigation had been implemented and that ChatGPT and Codex were recovering before confirming full recovery.
Anthropic also restored its affected Claude services, while Grok recovered somewhat later. xAI subsequently marked the model incident as resolved.

Azure connection remains unconfirmed
The near-simultaneous failures prompted scrutiny of possible shared infrastructure. Microsoft Azure emerged as one of the potential links, with reports indicating that Azure also experienced disruptions around the same period and that its infrastructure is used in AI workloads associated with the affected services.

However, neither OpenAI, Anthropic nor xAI has officially attributed the simultaneous outages to Azure. A direct causal link therefore remains unconfirmed.

Google’s Gemini also drew some user complaints during the period, although Google did not officially acknowledge a major outage. Gemini operates on Google’s own cloud infrastructure.

Technology researchers caution that major AI platforms typically operate across complex combinations of cloud infrastructure, data centres, networks and third-party services. A simultaneous outage therefore does not, by itself, establish a single point of failure.
Still, the incident underscores the growing dependence on AI services and the importance of resilient, diversified and independently recoverable infrastructure behind them.
//DBTech/EMO/IH/SME//

This is Adobe’s second acquisition in India, following its purchase of video platform Rephrase.ai in 2023. However, neither company has officially disclosed the financial terms of the deal.

Technology and Scope of Work

Relo had been developing technology to automate complex marketing processes using artificial intelligence. The technology enables companies to create, manage, and track AI-powered campaigns while maintaining their brand visibility across generative AI platforms such as ChatGPT, Gemini, and Claude.

As part of the acquisition, Relo’s six-member technical team will join Adobe’s core team. At the same time, Relo’s standalone operations will be completely discontinued, meaning its existing users will no longer be able to access the tool directly.

Relo’s Rise and Funding

Relo was founded in 2025 by two classmates from India’s prestigious IIT, Georgy Bobby and Dhruv Jaglan. Shortly after its launch, the company raised $1 million in funding at a valuation of $10 million from leading venture capital funds including Peak XV, DEVC, and Day Zero Ventures.

Since its launch, more than 10,000 users have used Relo’s workflow-related tools.

Market Competition and Strategic Impact

In recent times, technology giants such as Canva, Google, Meta, and Amazon have been strengthening their own AI-powered marketing systems. In response, Adobe has been taking a series of major steps to strengthen its marketing ecosystem, including the acquisition of search engine optimization platform Semrush for $1.9 billion last year.

According to analysts, Relo’s innovative AI workflows could play a significant role in helping Adobe’s enterprise customers create digital experiences and manage marketing activities.

//DBTech/BMT/OR//

At a press conference, New York Mayor Zohran Mamdani said the new decision would temporarily suspend the use of around 40 AI-powered educational software programs and apps currently used in the city’s classrooms.

He noted that direct human interaction with teachers and classmates is most important for children’s proper growth and learning. Students’ primary goal should be to develop their skills through interaction with peers, build positive relationships with teachers, and try to solve complex problems using their own intelligence when faced with challenges.

Under the new policy, approximately 600,000 public-school students through eighth grade will be directly restricted from using AI. However, high school students will be allowed to use AI technology under certain conditions in specific cases. Teachers, meanwhile, will continue to be allowed to use AI as before to assist with lesson planning, class schedules, and administrative tasks.

It is worth noting that New York City first temporarily banned ChatGPT shortly after it was launched in 2022. The order was later lifted, and specialized educational technology was introduced for use in schools.

However, the city has now decided on this comprehensive pause in an effort to reduce students’ excessive dependence on technology and maintain the quality of traditional education.

//DBTech/BMT/OR//

At least one in every five internet users aged 12 to 17 has faced such a horrifying experience. The study, conducted in countries across Asia, Africa, Latin America, and Eastern Europe, has highlighted major gaps in children’s social protection in the digital age.

The Trap of Social Media and Gaming

According to the report, around 15 million children have been exposed to unwanted sexual messages and inappropriate content. Children are also frequently being forced into sexual conversations or having their images shared without consent. Nearly 60 percent of these crimes occurred on popular social media platforms such as Facebook, WhatsApp, Instagram, Snapchat, and TikTok.

Online gaming platforms were used in approximately 14 percent of cases. Of particular concern, in more than half of the cases, the perpetrator was someone previously known to the child. However, due to social stigma and fear, fewer than 1 percent of children or families reported these incidents to the police or helplines.

AI Technology and Psychological Trauma

Artificial intelligence (AI) technology has now added a new dimension to the threat. According to data from nine surveyed countries, images of approximately 1.1 million children were manipulated using AI to create fake sexual images or videos.

Children subjected to such brutal online abuse are suffering severe psychological distress. The study found that affected children are at least four times more likely than other children to experience suicidal thoughts or tendencies toward self-harm.

Responsibility of the Tech Industry and the Way Forward

UNICEF Executive Director Catherine Russell said these incidents are causing deep psychological wounds in children, while most victims continue to suffer silently due to a lack of proper guidance.

Although technology giants such as Meta, Snapchat, and TikTok claim to have automated safety technologies and account-protection measures in place, such crimes have not stopped in practice. Against this backdrop, UNICEF has urged governments and technology companies to work together to strengthen protections for minors on digital platforms and to revise international laws in line with the rapid evolution of technology.

//DBTech/BMT/OR//

Uber CEO Dara Khosrowshahi announced the decision in an internal email sent to employees on Wednesday.

Management Restructuring and Scale of Layoffs

According to a report by Bloomberg, which first disclosed the information, the company is restructuring its overall administrative structure. As a result, the number of managers will be reduced by approximately 20 percent, with many of them being shifted to individual contributor roles.

In addition, the number of small teams consisting of only one or two members will be reduced by around 50 percent. Layoffs will also affect positions that are seven or more levels below the CEO.

Department Mergers and End of Remote Work

As part of the restructuring process, Uber is merging its engineering, science, and delivery divisions. At the same time, its restaurant, retail, and direct delivery services are being brought under a single umbrella.

The company is also making significant changes to its working arrangements, effectively ending the option of working from home or remotely. Going forward, fewer than 1 percent of employees will remain eligible for this arrangement.

Uber CEO’s Statement

In an email to employees, CEO Dara Khosrowshahi wrote, “We have launched new services, expanded the scope of our business, and become a much larger and stronger company than before. However, with this expansion, complexity within the organization has also increased—additional layers have been created and coordination has become more difficult. The administrative infrastructure that was appropriate on a smaller scale can no longer serve us effectively at our current size.”

//DBTech/BMT/OR//

After Bloomberg News first reported the incident, Dropbox officially acknowledged that the cyber intrusion took place between August 4 and August 21.

However, there is some relief: hackers were able to view or download confidential files from fewer than one-third of the affected accounts.

Password and Security Flaws

Following an initial investigation, Dropbox said the hackers primarily targeted accounts linked to technology company Lenovo ID that did not have two-factor authentication enabled.

By exploiting a technical flaw in an older integration framework of Lenovo ID, the cybercriminals were able to gain access without the main Dropbox password. Once the incident was detected, the authorities terminated all active Dropbox sessions operating through Lenovo ID.

At the same time, the connection between the two systems was severed, and users are now required to provide their own Dropbox password to regain access.

Legal Action and Impact on the Stock Market

Dropbox has already formally notified international data-protection regulators about the customer data security breach.

Meanwhile, Lenovo said in a statement that although some Dropbox accounts were affected because of the old technical vulnerability, none of Lenovo’s own customers were victims of the hacking incident.

As news of the cyberattack became public, Dropbox’s share price fell by approximately 2.4 percent on international stock markets. Affected users are already being urged via email to change their passwords and take additional security measures.

//DBTech/BMT/OR//

Debt Relief and Benefits for Shareholders

Under the terms of the agreement, GoPro shareholders will receive a total of $285 million in cash, equivalent to $1.14 per share. As a result, GoPro’s existing investors will retain approximately 10 percent ownership of the newly merged company.

In addition, once the deal is fully completed toward the end of this year, GoPro’s outstanding debt of approximately $92 million will be completely repaid. Despite the change in ownership, GoPro will remain listed on the Nasdaq stock exchange as before.

A New Direction: Artificial Intelligence and Military Technology

For the past several months, GoPro had been planning to expand beyond cameras and enter the military technology and data-center sectors. The newly formed Starman Optical primarily operates in optical-photonics technology. Through the merger, GoPro’s world-class lens and imaging technology intellectual property will be used to develop important equipment for AI data centers, robotics, aerospace, and military applications.

Charles Tebele, CEO of Starman Holdings, said the joint initiative was undertaken to bring the production of these critical technology products back to the United States.

The Future of Cameras and a Markiplier Surprise

Despite the transformation, GoPro is not leaving the action-camera market anytime soon. The company said it will continue offering its existing camera lineup and cloud subscription services to customers as before.

Interestingly, just before news of the deal became public, popular YouTuber Mark Fischbach, better known online as ‘Markiplier,’ purchased an 8.5 percent stake in GoPro, becoming the company’s largest single shareholder.

GoPro founder and CEO Nick Woodman hopes that through the deal, GoPro will establish a strong position as one of the United States’ leading imaging technology providers not only in the consumer market but also in the commercial and defense sectors.

//DBTech/BMT/OR//

The company’s research identified a phishing campaign in which more than 200,000 attacks used cloud platforms to steal Microsoft account credentials through a multi-stage deception process.

The campaign combines legitimate web services with fake CAPTCHA pages, dynamically generated web content and the Browser-in-the-Browser (BiTB) technique to make the attack harder to detect.

The attack begins by using the identity of a trusted or familiar person to persuade the victim to log in to a Microsoft account. In most cases, the victim receives a phishing link by email.

Clicking the link takes the user to a purported ‘anti-bot’ page, where the victim is asked to complete a fake CAPTCHA using a corporate email address.

Instead of verifying whether the user is human, the fake CAPTCHA collects the email address and redirects the user to a Cloudflare automatically assigned .workers.dev URL so the address can be passed to the next stage.

The victim is then taken to another page hosted on a free Cloudflare Workers subdomain, where a second CAPTCHA is displayed.

Although the second CAPTCHA is genuine, it is not directly embedded in the HTML code. Instead, it is dynamically added to the page, making the overall attack more difficult for security solutions to detect.

Fake Office 365 login window

The final stage presents the victim with a fake login window designed to resemble Office 365.

Created using the Browser-in-the-Browser (BiTB) technique, the window contains an address bar and window controls designed to look like those of a genuine browser.

However, it is actually a floating window rather than a real browser window.

When users enter their usernames, passwords and multi-factor authentication (MFA) codes, the information is transmitted to the attackers.

The malicious script also collects the victim’s session cookies. Once the information has been captured, the victim is redirected to a generic error page, making it less obvious that a phishing attack has taken place.

Legitimate cloud services become attack infrastructure

Kaspersky cybersecurity expert Olga Altukhova said, “Attackers actively exploit legitimate services due to their reputation, free plans, and tools that they can exploit. What’s more, in the example that we investigated in the report, phishers were able to create a multi-stage Adversary-in-the-Middle attack, proxying all traffic from what looked like a legitimate Microsoft website and combining it with Browser-in-the-Browser techniques. This shows how phishing techniques are becoming more and more sophisticated,” 

Her assessment indicates that identifying suspicious or obscure websites alone is no longer sufficient to detect phishing campaigns.

Because attackers can operate through legitimate and familiar cloud infrastructure, security teams increasingly need to examine webpage behaviour, scripts and dynamically generated elements alongside domain reputation and digital certificates.

Kaspersky’s five safety recommendations

Kaspersky has advised users to follow several precautions.

First, be cautious when dealing with CAPTCHA. CAPTCHA mechanisms generally do not request personal information such as an email address. If a verification process asks for personal information to prove that the user is not a robot, it should be treated as a potential warning sign of phishing or fraud.

Second, be cautious about unexpected login requests. A request should not automatically be considered safe simply because it comes from a trusted domain or uses a valid SSL certificate.

Third, check the URL in the actual browser address bar. In a Browser-in-the-Browser attack, criminals can imitate an address bar or popup inside a fake window. They cannot, however, change the domain displayed in the browser’s genuine address bar. Users should therefore verify the URL in the actual browser interface.

Fourth, keep browsers and security extensions updated. Outdated software and security components may have reduced capabilities against newer attack techniques.

Fifth, do not rely solely on domain reputation. Kaspersky recommends using reliable cybersecurity solutions capable of analysing webpage scripts and dynamic elements in addition to checking the reputation of the domain.

The campaign highlights how phishing operations are evolving beyond conventional fake websites. By combining legitimate cloud infrastructure, dynamic web technologies and convincing browser-interface replicas, attackers are attempting to build user trust while simultaneously making their campaigns harder for conventional security controls to identify.

//DBTech/RICP/SMP//

Although the brand is best known for its portable and Bluetooth party speakers, this marks its first major move toward a ‘whole-home audio’ system. The company has announced three different models under the new series.

Seamless Connectivity and Soundbar Integration

The new Cove series speakers can be easily connected to other speakers in any room of the house through the ‘JBL One’ app. This allows users to enjoy uninterrupted music throughout every corner of their home without losing connectivity or having to change the settings again.

In addition, to make the movie-watching experience more immersive, the speakers can be connected as rear speakers to soundbars from the ‘JBL Bar 2’ series, creating a modern theater-like atmosphere at home.

Technology and Prices of the Three Models

The first model in the series, the ‘Cove M1’, is designed specifically for everyday music listening. Featuring three tweeters and one woofer, the speaker uses AI-powered Sound Boost technology and is priced at $230.

The second model is the ‘Cove P1’, a portable, battery-powered speaker. Its powerful battery can provide up to around 14 hours of playback on a single charge, while Dolby Atmos technology ensures impressive sound quality. The speaker is priced at $400.

The most powerful and flagship model in the series is the ‘Cove X1’. Equipped with four tweeters and two woofers, the device uses Dolby Atmos technology to deliver a clear and deep sound experience throughout the room. This premium speaker is priced at $450.

The Cove series speakers will be available for purchase from September 13 in two attractive colors—black and white. The flexibility to purchase speakers one at a time and gradually build a whole-home audio system according to individual needs is being viewed as a new offering for audio enthusiasts.

//DBTech/BMT/OR//

The bilateral agreement was reached to secure access to infrastructure at a state-of-the-art data center that is being developed in Nueces County, Texas.

A source directly familiar with the matter confirmed the information to international news agency Reuters.

Technology and Infrastructure Capacity

The new data center project in Texas, with a capacity of 350 megawatts, is being jointly developed by Hut 8, a company that has transformed itself from a cryptocurrency mining operation into an AI infrastructure provider.

Under the agreement, Anthropic will obtain additional cloud computing capacity from Nvidia-backed Lambda to meet the rapidly growing demand for its artificial intelligence chatbot Claude and programming tool Claude Code.

Anthropic is also planning to go public through an initial public offering (IPO) in the near future. With that goal in mind, the company has been continuously expanding its computing capacity.

Competition for AI Investment

Since the global rise of generative AI technology, demand for computing infrastructure and high-capacity data centers has increased dramatically. Leading semiconductor and AI chipmaker Nvidia is playing a key role as one of the catalysts behind this deal.

According to a report by The Wall Street Journal, Nvidia itself is playing a major role in securing the lease for the main data center. Earlier last week, Anthropic also announced a massive $45 billion investment to obtain AI cloud services from a data center campus in West Virginia.

According to technology analysts, the infrastructure race among AI giants is becoming increasingly long-term as companies seek to strengthen their positions in the global market.

//DBTech/BMT/OR//

The service was initially launched in the United States, but the Japanese technology giant plans to expand it to other regions, including Canada, very soon.

A Vast Content Library

Sony has named the new service Live TV on PS5. Users can access it directly through the console’s dedicated Media tab.

Alongside original content from Sony Pictures Television, the service will offer movies, dramas, anime, sports, news and documentaries from popular platforms such as Lionsgate and Crunchyroll. Viewers will be able to watch popular shows including The Walking Dead and Community, as well as well-known anime series such as Naruto, Death Note and Sailor Moon, at no cost.

A New Business Strategy

According to market analysts, Sony’s initiative is aimed at retaining existing customers within its own ecosystem rather than relying solely on sales of new gaming consoles.

By creating a new advertising-based revenue stream within its own ecosystem, Sony is also seeking to prevent users from turning to third-party free streaming services such as Roku or Tubi.

The ability to switch directly to live television during a break from gaming without turning off the console is being seen by technology enthusiasts as a new boost to the PS5’s appeal.

//DBTech/BMT/OR//

Expansion into International Markets

Following its success in the U.S. market, OpenAI has announced that its ChatGPT Ads Manager service will now be made available to advertisers in India, Europe, the Middle East and North Africa. As a result, businesses in these regions will be able to promote their products directly on ChatGPT without relying on intermediaries.

OpenAI said a significant portion of its advertising revenue comes from small and medium-sized businesses around the world. At present, these advertisements are being shown to users of the free version of ChatGPT as well as the lower-cost Go plan.

Targets and Global Competition

Although OpenAI’s rapid growth is being viewed as a major achievement, market analysts believe the company still faces several challenges. Technology research firm eMarketer said that reaching a $1 billion annualized run rate in such a short period is impressive, but the pace is somewhat slower than OpenAI’s own target of $2.5 billion for 2026. To reach that goal, the company will need to accelerate advertising revenue growth during the remainder of the year.

To establish a stronger position in the global online advertising market, OpenAI is facing direct competition from technology giants such as Google, controlled by Alphabet, and Meta, the parent company of Facebook.

Alongside making artificial intelligence accessible to users worldwide, the U.S.-based AI company is focusing on building a sustainable and diversified revenue model combining subscriptions, API services and advertising income.

//DBTech/BMT/OR//

The joint lawsuit was filed in a federal court in Washington state, with 22 U.S. states joining the FTC as co-plaintiffs.

The lawsuit alleges that since 2019, Amazon has manipulated auctions for advertisements displayed on its search pages, secretly collecting more than $20 billion in additional payments from advertisers without their knowledge.

Alleged Secret Pricing Tactics

According to the lawsuit, Amazon violated the standard rules governing auctions for advertisements displayed when users search for products on its platform. Under the common practice in digital advertising, the price is generally determined based on the bid submitted by the second-highest bidder. However, Amazon allegedly interfered with around 80 percent of its auctions and charged winning advertisers the full amount of their highest bids.

The FTC claims that this secret practice severely harmed at least 1.2 million businesses and small and large sellers that regularly promote their products on the platform. It further alleges that higher advertising costs ultimately forced ordinary consumers to pay more for products.

Amazon Denies Allegations

The Seattle-based technology company has formally denied all of the regulator’s allegations. In a statement, Amazon said its auction system is highly competitive and that product relevance, rather than bid price alone, receives the greatest consideration when selecting winning advertisements.

Amazon also disputed the FTC’s claims, saying the average cost per advertising click did not increase between 2019 and 2024. The company further claimed that its technology helped advertisers save around $8 billion between 2021 and 2025.

Amazon currently controls the world’s third-largest digital advertising market, behind Google and Meta. However, following news of the latest legal challenge, the company’s shares fell by around 2.5 percent in the stock market.

//DBTech/BMT/OR//

How the System Works

A commercial Wi-Fi extender typically costs around $50. However, an Android phone can perform a similar function completely free of charge. The phone remains connected to the main Wi-Fi router, receives its signal and then broadcasts that connection through a separate hotspot network of its own.

As a result, laptops, tablets or smart TVs can connect to the phone’s hotspot instead of directly connecting to the main router and access the internet. Although it does not perform like a mesh network, it can provide a quick solution in areas with weak Wi-Fi signals.

Simple Setup Process

1. First, fully charge the old phone and update it if necessary, then connect it to your home’s main Wi-Fi network.
2. Go to Settings > Hotspot and Tethering on the phone and turn on the Wi-Fi Hotspot option.
3. Change the hotspot name to one of your choice and set a strong password.
4. If the phone supports it, you can select the 5GHz band, which provides better speeds for nearby devices. For longer distances, however, 2.4GHz is generally more effective.
5. Place the phone roughly halfway between the main router and the Wi-Fi dead zone. Keep it plugged into a power outlet, as continuously running a hotspot can drain the battery quickly.//DBTech/BMT/OR//

The two Japanese companies produce more than three-quarters of all vehicles manufactured in Canada. If the new tariff structure takes effect from January 1 next year, they could be forced to shut down some of their assembly plants in the country.

Serious Risk of Production Shutdowns

The impact of the tariffs on Japanese automakers will be particularly severe in the U.S. market. According to international analyst firm Barclays, nearly one-quarter of Honda’s total U.S. sales last year and 17 percent of Toyota’s sales came from vehicles produced at Canadian plants. If the existing 25 percent tariff is doubled to 50 percent, it could cause a major disruption to the Japanese companies’ businesses.

Japanese automakers are already facing intense competition from cheaper Chinese electric vehicles in markets across Southeast Asia, Europe and Latin America. If access to their largest market—the United States—becomes more difficult for Canadian-made vehicles, the situation could become even more complicated.

Market Restructuring and Major U.S. Investment

Canada’s automotive industry produces around 1.2 million vehicles annually and directly and indirectly supports more than 427,000 jobs. Toyota’s RAV4 and Honda’s CR-V, two popular SUV models among the best-selling vehicles in the U.S., are largely manufactured at plants in Canada.

Faced with Trump’s unpredictable trade policies, Toyota has been compelled to formulate a five-year, $10 billion investment plan to expand production directly in the United States. As part of the plan, the company is building a new plant in Texas at a cost of $3.6 billion.

Free Trade Agreement Faces Uncertainty

Meanwhile, Honda is struggling to revive its loss-making automotive business. Uncertainty also remains over the USMCA free trade agreement between the United States, Canada and Mexico. The company has indicated that it may suspend plans to build a new plant in North America if the agreement is not renewed.

If the new tariffs come into effect, the two companies could be forced to send vehicles manufactured in Canada to markets outside the United States. However, these vehicles are specifically designed for the U.S. market, making it extremely difficult to suddenly redirect them to other countries or rapidly increase production capacity at plants elsewhere.

As a result, despite the extreme uncertainty, the Japanese automakers are currently monitoring the situation while awaiting a final decision.

//DBTech/BMT/OR//

SB Energy, which develops renewable-energy projects and data-center infrastructure, is preparing to go public through an initial public offering (IPO). The company could receive a valuation of more than $50 billion when it enters the public market.

The disclosure of OpenAI’s massive warrant package has attracted significant attention across the technology sector, highlighting the increasingly close relationship between artificial intelligence and energy infrastructure.

OpenAI, Nvidia and SoftBank Join Forces

OpenAI is not the only major technology company closely tied to SB Energy. Semiconductor giant Nvidia recently invested $1.5 billion directly in the company.

Nvidia has also reportedly agreed to provide up to $10.5 billion in financing support to help OpenAI lease a massive SB Energy data-center campus in Ohio.

Under their broader arrangements, SoftBank and OpenAI will jointly serve as long-term tenants at SB Energy’s three major data-center campuses.

Energy Infrastructure for the AI Boom

Founded in 2019, SB Energy develops and operates large-scale solar and wind power projects as well as data-center infrastructure. The enormous computing requirements of AI models and chatbots have created rapidly growing demand for reliable electricity and sophisticated data centers.

To meet this demand, SB Energy is developing several new renewable-energy-powered data-center campuses.

The combined involvement of OpenAI, Nvidia and SoftBank in financing and developing SB Energy points to a broader strategy in the AI industry: securing the energy and physical infrastructure needed to support the next generation of large-scale AI computing.

//DBTech/BMT/OR//

The newly launched observatory can capture images of a region of space at least 100 times larger than the area observed by the Hubble Space Telescope in a single exposure.

Shortly after launch, scientists confirmed that the telescope’s solar panels and sunshield had been successfully deployed. Over the coming weeks, its other technological systems will be activated one by one.

Once fully operational, Roman is expected to transmit around 1.4 terabytes of data to Earth every day. However, the public will have to wait until early 2027 to see its first images.

A New Window into the Universe

Highlighting the mission’s significance, Nicky Fox, head of NASA’s Science Mission Directorate, said Roman would open a new door to discovery and bring scientists closer to answering some of the deepest questions about cosmic history.

With its exceptionally wide field of view and rapid survey capabilities, the telescope will help reveal previously unseen regions of the universe. Its observations could also contribute to the search for planets beyond our solar system and help scientists understand the conditions under which planetary systems form.

Roman will investigate dark energy, the mysterious force associated with the accelerating expansion of the universe, as well as dark matter. It will also observe new planets and the earliest stages of planet formation.

Scientists expect the telescope to analyze light from roughly one billion galaxies over the course of its mission, potentially providing unprecedented insights into the structure, evolution and ultimate fate of the universe.

//DBTech/BMT/OR//

CXMT and Xiaomi recently jointly confirmed the new agreement on social media platform Weibo. The development comes less than a year after CXMT began mass-producing its previous-generation LPDDR5 chips.

Technological Progress and Soaring Revenue

Designed for mobile devices, the LPDDR6 is CXMT’s most advanced dynamic random-access memory (DRAM) chip to date. The breakthrough has significantly narrowed the gap between the Chinese manufacturer and global memory-chip giants such as Samsung and SK Hynix.

Xiaomi has said its new foldable phone will feature its own 3-nanometer ‘Xring O3’ processor, which is directly compatible with the new memory technology.

CXMT has also reported a major financial breakthrough in its first half-year report since being listed on the stock market. Compared with the previous year, the company’s revenue surged by 874 percent to 150.3 billion yuan, or approximately $22.36 billion.

Legal Battle and International Challenges

The Hefei-based chipmaker has already begun sending samples of its new chips to customers for use in smartphones, servers and smart vehicles. However, alongside its commercial success, CXMT is facing a significant international challenge.

The U.S. Department of Defense, or Pentagon, has placed the company on a special blacklist over allegations that it supports China’s military. CXMT filed a lawsuit in a U.S. court last Friday challenging the Pentagon’s decision.

Although the Pentagon has declined to comment on the ongoing legal proceedings, CXMT has said it will continue its legal battle in an effort to strengthen its position in the global chip market.

//DBTech/BMT/OR//

Technology platform WABetaInfo was the first to identify the new feature in WhatsApp Android beta version 2.26.34.1. The technology, which has been under testing, is expected to be officially rolled out to all users soon.

How Scam Alert Works

The primary purpose of the new feature is to immediately warn users when they receive suspicious or potentially fraudulent messages from an unfamiliar or unknown number. A key advantage is that the sender will not be notified that the recipient has received the warning or has the Scam Alert feature enabled.

When an alert appears for a particular message, users will have the option to block and report the sender to WhatsApp or mark the message as safe and continue communicating with the person.

If a user marks a chat as trusted, WhatsApp will optionally allow the user to send the last five messages from that conversation to the company to help improve the system’s performance and accuracy. Users will have to enable this option themselves by going to the Account section in the app’s settings.

On-Device Data Analysis and Strong Privacy

According to an official statement from Meta, WhatsApp’s new security system will operate entirely through a machine-learning model running on the user’s device.

Information from messages will not leave the user’s phone and be sent to Meta, WhatsApp or any third party for analysis. As a result, WhatsApp’s end-to-end encryption will remain intact, while users’ private conversations and personal information will remain protected.

Experts believe that using artificial intelligence directly on users’ devices to identify suspicious activity could play an important role in protecting people from scammers without compromising their privacy.

//DBTech/BMT/OR//

The joint lawsuit, filed in the US District Court for the Northern District of California, alleges that Anthropic carried out an unprecedented campaign to obtain thousands of copyrighted songs and lyrics illegally through torrents, website scraping and downloads. The major legal action by the music industry was first reported by Music Business Worldwide.

In the lawsuit, the publishers claim that Anthropic used thousands of songs and lyrics without permission to train its popular AI model Claude, which they describe as an outright act of theft.

Facing the allegations, however, Anthropic has said it will strongly defend its position. In an email to technology news outlet TechCrunch, an Anthropic spokesperson said the company disagrees with the publishers' allegations and intends to mount a vigorous legal defense in court.

Not Anthropic's First Copyright Battle

This is not the first intellectual property lawsuit brought against Anthropic. The same lawyers involved in lawsuits filed by Concord Music Group and Universal Music Group, as well as the high-profile Bartz v. Anthropic case concerning authors' rights, are representing Sony and Warner in the new lawsuit.

In the landmark Bartz case, a judge ruled that although copyrighted material may in some circumstances be used for AI training, obtaining such material through piracy or other unlawful means is illegal. Anthropic ultimately agreed to settle that case for $1.5 billion.

Alleged Use of Torrent Networks

The latest lawsuit is broader in scope than Anthropic's previous legal battles. The publishers directly accuse the company of using illegal torrent networks to obtain copies of sheet music containing vast quantities of books, songs and lyrics.

The publishers therefore allege that Anthropic engaged in widespread piracy to acquire copyrighted material for its AI systems.

According to technology analysts, the collective legal action by the music publishers signals a tougher approach toward how data is collected for AI training. The case could potentially create significant challenges for Anthropic's business operations and further shape the legal boundaries surrounding AI companies' use of copyrighted material.

//DBTech/BMT/OR//

Although neither computer features major changes to its exterior design, Apple has made significant internal upgrades to improve processing performance. Both computers require separate keyboards, mice and monitors. However, there is a substantial difference between the two in terms of performance and price.

Mac Mini for Everyday Users and Professionals

The Mac Mini is a compact desktop computer designed for everyday use and general productivity. Measuring just 2 inches in height, the sleek computer is powered by Apple's new M6 chip, manufactured using a 2-nanometer process.

The base model comes with a 12-core CPU, 12-core GPU, 16GB of RAM and 256GB of SSD storage. It is suitable for everything from everyday tasks to entry-level video editing.

For more demanding workloads, Apple is also offering a professional version of the Mac Mini powered by the M5 Pro chip. It features an 18-core CPU, 16-core GPU, up to 64GB of RAM and up to 8TB of SSD storage. This configuration is particularly well suited to 3D graphics and coding.

The starting price of the new Mac Mini has been set at $899, somewhat higher than that of its previous generation.

Mac Studio for Power Users and Heavy Workloads

For highly complex and demanding workloads, Apple has introduced the Mac Studio. Standing 3.7 inches tall and weighing nearly 2.75 kilograms, the substantial desktop computer is powered by the high-performance M5 Max chip.

Its base configuration features an 18-core CPU and 32-core GPU, allowing it to handle demanding tasks ranging from developing artificial intelligence models to intensive gaming.

For maximum performance, Apple is also offering a flagship version powered by the M5 Ultra chip. Built by combining two M5 Max processors, the chip features a 36-core CPU and 80-core GPU, with support for up to 512GB of RAM and 16TB of storage. The system is primarily designed for animation studios as well as large-scale engineering and research work.

For connectivity, the Mac Studio includes four Thunderbolt 5 ports, two USB-A ports, 10-gigabit Ethernet and two USB-C ports on the front.

The Mac Studio starts at $2,499.

//DBTech/BMT/OR//

The feature will allow users to read the information contained in a podcast without turning on the sound while scrolling through their feeds. Threads head Connor Hayes recently confirmed the experimental feature in a post of his own.

How the Technology Will Work

Under the new system, whenever a podcast or audio file is posted on Threads, the speaker’s words will appear on the screen as text or subtitles in real time.

Connor Hayes said the feature is just one part of several new tools Threads is developing to support podcast creators.

The feature could be particularly useful in crowded places or other situations where playing audio aloud is inconvenient. Since Threads is primarily a text-based social network, adding written content alongside audio could also fit naturally with the platform’s overall experience, technology analysts say.

Long-Term Plan to Attract Podcasters

Audio and video creators are becoming increasingly important across digital platforms. Since last year, Meta has been introducing various tools to promote podcasts on its text-focused social network.

In June this year, Threads surpassed 500 million monthly active users. With such a large potential audience, the company considers specialized features for audio creators increasingly important to retaining podcasters and encouraging them to publish on the platform.

Although it is still unclear whether the experimental feature will eventually become available to all users, Threads’ growing focus on attracting content creators is being viewed positively.

//DBTech/BMT/OR//

The smartwatch received a score of 8.9 out of 10 in a review by technology news outlet Engadget. Rather than overwhelming users with unnecessary features, the device focuses on essential functions such as reading notifications, controlling music and checking the weather—and handles them efficiently.

Design and Display

The Pebble Time 2 features a 1.5-inch, 64-color always-on e-paper touchscreen display protected by Gorilla Glass.

Its CNC-machined stainless-steel body and tactile physical buttons give the device a sturdy and premium appearance. The e-paper display remains highly visible even under bright sunlight. The watch also supports any standard 22mm strap, making it easy to customize.

Improved Performance and Long-Lasting Battery

Although the watch uses a relatively modest processor, its lightweight PebbleOS software allows it to run smoothly without noticeable lag or sluggishness.

The built-in 185mAh battery can easily last up to nine days under normal use, while a power-saving mode can extend battery life to around 20 days.

For convenient charging while travelling, the watch comes with a dedicated USB-C adapter. It can also track basic physical activities, including heart rate, walking steps and sleep.

Open-Source Software and Price

The smartwatch runs on fully open-source software and offers thousands of third-party watch faces and apps. However, some basic features, such as a stopwatch and timer, require users to download third-party apps from the app store.

Priced at $225 in international markets, the device may not appeal to everyone. But for users looking for a simple, long-lasting smartwatch rather than an overly complicated piece of technology, the Pebble Time 2 could be an excellent choice.

DBTech/BMT/OR

Several ChatGPT capabilities are now available for free or through comparatively lower-cost plans, making it harder for some users to justify continuing a monthly payment.

In addition, OpenAI's recent changes to its ChatGPT model lineup, including the retirement of the popular GPT-4o model, have caused dissatisfaction among some users. Policy debates surrounding OpenAI's agreements with the US Department of Defense have also prompted some people to reconsider their use of the service.

The Correct Way to Cancel Your Subscription

The cancellation procedure primarily depends on where the subscription was purchased. Users who subscribed directly through ChatGPT's website can cancel it from their ChatGPT account. The current OpenAI Help Center instructs web subscribers to open their account menu, go to Settings, then Account, and select the relevant cancellation option. Users should cancel at least 24 hours before the next billing date to avoid being charged for another subscription period.

Users who subscribed through an iPhone or Android app must cancel the subscription through the Apple App Store or Google Play, respectively. Simply deleting the ChatGPT app from a smartphone does not cancel the subscription or stop future charges.

Data Protection and Account Deletion

Cancelling a paid subscription does not automatically delete the user's ChatGPT account or existing conversation history.

Users who do not want their new conversations to be used to improve OpenAI's models can turn off “Improve the model for everyone” under Settings → Data Controls. Their conversations can remain in their chat history even after this setting is disabled.

However, anyone wishing to permanently remove their ChatGPT account and associated data must separately request account deletion. Account deletion is permanent and cannot be undone. OpenAI says it generally deletes account data within 30 days, although some limited information may be retained for longer when required or permitted by law.

Users with subscriptions purchased through Apple or Google should also remember that deleting the OpenAI account does not cancel those mobile subscriptions. They must cancel the subscription through the relevant app store to prevent further charges.

Refunds and Remaining Subscription Period

Cancelling a subscription does not normally result in an automatic refund of previous payments. Instead, cancellation stops future renewals, while paid features generally remain available until the end of the current billing period.

For accidental purchases, users can generally request a refund if they contact OpenAI within 14 days of the charge. For subscriptions purchased through ChatGPT or Google Play, refund requests are handled through OpenAI Support; Apple App Store purchases require users to request refunds directly from Apple.

Users in the European Union, United Kingdom or Turkey can receive a prorated refund if they cancel within 14 days of purchase, subject to the applicable conditions.

DBTech/BMT/OR//

Smartphones that once took a long time to charge using conventional cables can now recharge remarkably quickly. Almost every major technology company has also developed its own charging system. However, simply seeing a USB Type-C port on a device does not guarantee that it supports fast charging. Charging speed primarily depends on the device's internal hardware and charging architecture.

Major Technologies and Proprietary Charging Systems

The most popular and universal charging standard for smartphones, tablets and laptops today is USB Power Delivery (USB-PD). Flagship devices such as Apple's iPhones, Samsung's Galaxy series and Google's Pixel lineup use this technology.

An advanced version of USB-PD is Programmable Power Supply (PPS). It continuously adjusts the voltage and current supplied to a phone according to the battery's requirements while charging. This helps manage heat and can make charging more efficient and safer.

Meanwhile, Qualcomm's Quick Charge, Oppo's SuperVOOC and Xiaomi's HyperCharge are proprietary charging technologies developed by individual brands. To achieve their maximum charging speeds, compatible devices generally require the specific charger and cable supported by the respective manufacturer.

It is also important to understand that a phone will not automatically accept more power simply because a higher-wattage charger is connected. For example, if a phone supports a maximum charging capacity of 20 watts, connecting a powerful 100-watt charger will not force the phone to draw 100 watts. The device regulates the amount of power it accepts according to its own capabilities.

Battery Life and Safety Measures

Many users believe that fast charging reduces a smartphone battery's lifespan. Since rapid charging can generate additional heat, there is some concern that prolonged exposure to high temperatures could accelerate degradation of lithium-ion batteries.

Modern smartphones, however, incorporate sophisticated automatic temperature and power-management systems. As the battery approaches full capacity, the phone can automatically reduce the charging rate to control heat and stress on the battery.

In addition, features such as Apple's Optimized Battery Charging and Battery Protection options on Samsung and Google devices can limit charging to around 80 percent under certain settings and circumstances, helping extend long-term battery health.

According to technology analysts, users looking for a single charger that can serve multiple types of devices should consider a high-wattage charger supporting both USB-PD and PPS. Such a charger offers broad compatibility while allowing supported devices to determine the appropriate charging speed automatically.

//DBTech/BMT/OR//

Discussions focused on integrating spatial technology, land degradation neutrality, renewable energy systems, and climate adaptation frameworks. Minister Abdul Awal Mintu stated, "Addressing cross-border environmental disruptions requires robust regional and global technology-sharing partnerships. Bangladesh aims to share its localized ecosystem management practices while adopting advanced technological interventions from partner nations."

During talks with Indian Environment Minister Bhupender Yadav, both sides prioritized deploying remote sensing and Geographic Information System (GIS) tools for Sundarbans coastal monitoring, alongside technical exchanges in battery energy storage systems (BESS) and land-efficient solar installations.

In discussions with Mongolian Environment Minister Sandag-Ochir Tsend, both delegates agreed to draft a comprehensive Memorandum of Understanding (MoU) targeting pasture management, drought early-warning systems, and joint pilot research. Additional sessions with Mongolian Foreign Minister Battsetseg Batmunkh addressed bilateral trade and skilled labor mobility, while talks with IOM Deputy Director General Ugochi Daniels centered on climate-induced migration tracking and vulnerable community support frameworks.
//DBTech/THEP/NAP/MIS//

In a devastating turn of events, a catastrophic flash flood in the Rasuwa district of Nepal swept away mountain settlements like matchsticks, leaving behind only debris and grieving families. Following this geographical tragedy, discussions surrounding the 2009 blockbuster sci-fi movie ‘2012’ have resurfaced globally across digital news platforms and social media networks. The film, starring John Cusack, Amanda Peet, and Chiwetel Ejiofor, dramatically portrayed catastrophic floods submerging the Himalayas. The striking visual resemblance between the real-life disaster and the 15-year-old fictional film has sparked intense online debate.

However, beyond ‘2012’, an analysis grounded in science, engineering, and climate technology reveals that several other sci-fi movies have long foreshadowed these environmental and geographical crises.

Science and Cinema: Coincidence or Environmental Warning?
Directed by Roland Emmerich, '2012' depicted a apocalyptic scenario triggered by Earth’s crustal displacement and solar radiation anomalies. One of its most memorable sequences showed massive waves surging over the Himalayan peaks as a Buddhist monk rang a monastery bell. As recent video clips of cloudbursts and flash floods in Nepal went viral online, netizens drew parallels to the iconic movie tagline: "We Were Warned."

While meteorologists and climate scientists view cinematic prophecies as fictional dramatizations, the underlying scientific reality holds true. Accelerated global warming, coupled with unviable engineering infrastructure in fragile ecological zones, has heightened the risk of Glacial Lake Outburst Floods (GLOF). What was portrayed on screen as a sudden cinematic doom is occurring in reality through measurable ecological degradation.

Parallel Depictions in Other Sci-Fi Movies
The Day After Tomorrow (2004): This science-driven film illustrated rapid atmospheric shifts triggering superstorms and flash deluges across global mountain systems. The narrative closely parallels current online data tracking unpredictable cloudburst events in South Asian high-altitude regions.
Geostorm (2017): Highlighting satellite technology and climate engineering systems, the film shows how human error and tech manipulation lead to catastrophic global flooding. It serves as a stark allegory for unchecked engineering interventions in fragile ecosystems like the Himalayas.
Interstellar (2014): Christopher Nolan’s masterpiece explores extreme environmental collapse, reminding modern technological societies of the consequences of failing to protect Earth's ecosystems.

Coincidence vs. Scientific Reality
Analysts maintain that sci-fi films do not possess supernatural foresight; rather, screenwriters extrapolate real scientific models and geographic vulnerabilities into visual narratives. The catastrophe in Rasuwa, Nepal, underscores the severe costs of ignoring climate science data.

The viral comparison of '2012' clips with real-life footage is no longer just a movie trivia topic online. It stands as a stark reminder of the urgent need for sustainable engineering, climate science integration, and responsible ecological management in the Himalayan region.
//DBTech/IH/MIS/SMP//

In a statement issued on Thursday, 27 August 2026, the embassy announced that concerned parties can access the emergency hotlines managed by the Counsellor & Head of Chancery and the First Secretary (Labour Welfare Wing). The mission is actively coordinating with local Russian administrative and rescue agencies to track the missing and secure emergency medical interventions for the wounded.

Diplomatic and engineering sources indicated that the chemical infrastructure—located approximately 7,000 kilometers from Moscow—was being operated under China's Sinopec Engineering Group. The incident occurred when an active gas trunk line ruptured during high-pressure diagnostic testing. The ensuing fire rapidly consumed the system's pipeline check valve, triggering a massive industrial explosion.

The accident resulted in injuries to an estimated 50 to 60 multinational technical personnel from Russia, China, Bangladesh, Nepal, Azerbaijan, Uzbekistan, and Tajikistan. Search and rescue operations are ongoing amid heavy structural debris and dense combustion smoke.
//DBTech/PR/MIS//

Discussions centered on deploying modern nuclear technology, Small Modular Reactors (SMR), grid reliability, clean energy development, and mutual knowledge-transfer frameworks between both nations.

Minister Fakir Mahbub Anam emphasized that adopting safe, modern nuclear engineering solutions is vital to address rising national power demands and maintain long-term energy security. He stated, "Bangladesh is actively exploring the feasibility of advanced nuclear systems, especially Small Modular Reactors. We remain committed to the peaceful, secure, and responsible deployment of nuclear technology. Expanding technical synergy, engineering expertise, and bilateral partnerships with advanced nations like the United States is critical to accelerating economic growth."

US Secretary of Energy Chris Wright expressed a positive stance toward fostering nuclear and energy cooperation with Bangladesh. He highlighted the strategic priorities of the US Department of Energy in advancing research, engineering deployment, and commercial scaling of cutting-edge SMR technologies globally.

Both delegates reaffirmed their commitment to deepen bilateral relations and explore joint engineering initiatives within nuclear science, energy systems, and modern technology.
//DBTech/MAM/SME//

In an official filing with the US Securities and Exchange Commission (SEC), the company confirmed that access to its IT systems and critical business applications had been severely disrupted.

The Massachusetts-based medical-device giant is globally known for producing sensitive healthcare equipment, including pacemakers and defibrillators. The incident has therefore raised new concerns about cybersecurity across the healthcare technology sector.

Disrupted Product Shipments and Office Chaos

The cyberattack has disrupted Boston Scientific’s normal processes for accepting new orders and shipping medical equipment around the world. After the company’s internal network systems became unavailable, thousands of employees working at its Cork campus in Ireland were told to stop working and sent home.

The company’s corporate infrastructure relies heavily on Microsoft and Amazon Web Services. However, the company has yet to determine exactly how the attack began or who was responsible. It has also been unable to provide a specific timeline for restoring the affected systems.

Concerns Over Patient Safety and Limited Information

Around 48 million patients worldwide rely on various Boston Scientific medical devices each year. However, it remains unclear whether the cyberattack has affected the functionality of implanted devices or directly compromised patient safety.

In comments to technology news outlet TechCrunch, company spokesperson Shenelle Hastings only reiterated the company’s official statement. She declined to directly answer questions about whether patients needed to take any special precautions or what type of cyberattack had occurred.

Growing Cyber Threat to the Healthcare Sector

A growing number of major healthcare technology companies around the world have recently been targeted in a series of cyberattacks. The Boston Scientific incident is the latest addition to this worrying trend.

Earlier this year, global medical-device manufacturers Abbott Laboratories and Medtronic were also targeted by hackers. Previously, Iran-backed hackers carried out a major cyberattack against US company Stryker. In that incident, the attackers exploited Microsoft’s centralised device-management portal to remotely wipe thousands of devices used by Stryker employees across its Windows network.

The repeated attacks have raised serious questions about the overall security of the healthcare sector worldwide.

//DBTech/BMT/OR//

The change is primarily aimed at intensifying competition with other text-focused social-media platforms and expanding opportunities for users to create and share video content.

New Video Duration and Features

Alongside increasing the maximum video duration, Bluesky has also raised the maximum file-size limit. Under the new rules, users can now upload video files of up to 300 megabytes (MB). The company has also confirmed that video-upload speeds have been increased by two to three times compared with the previous system.

Bluesky’s new 10-minute limit is the same as the duration of videos recorded directly through the popular video app TikTok. In addition, longer videos can now be posted on other applications built on the same decentralised ‘AT Protocol’ infrastructure as Bluesky, including Bluescreen, Reelo and Skylight.

The Competitive Landscape

In recent months, almost every major social-media platform has been placing greater emphasis on video services. Meta-owned platform Threads, for example, allows users to upload videos of up to five minutes.

Meanwhile, users of the microblogging platform X, formerly Twitter, can post videos of up to 140 seconds for free. Paid users, however, can upload files of up to 16 gigabytes (GB) and videos lasting between three and four hours.

X has also recently introduced a new video editor to encourage users to upload original content.

Efforts to Expand Its User Base

Through the major expansion of its video service, Bluesky is primarily seeking to increase the number of active users on its platform. According to data from global market research and analytics firm Similarweb, Bluesky’s annual active-user base had declined by more than 25 percent by July this year.

Against this backdrop, technology analysts believe the longer video limit and faster upload system could play an important role in attracting new content creators to the platform.

//DBTech/BMT/OR//

Under the comprehensive binding agreement, Meta committed to executing structural changes across its platform algorithms. The reforms mandate enforceable daily usage limits for teen accounts on Instagram and Facebook, automated night-time feature restrictions, and enhanced algorithmic parental controls across the United States. While consenting to the financial resolution and policy restructuring, the Menlo Park-based conglomerate denied any legal wrongdoing. Meta’s shares gained 4.4 percent in pre-market trading following the announcement.

Court filings alleged that Meta systematically breached the Children's Online Privacy Protection Act (COPPA). Investigators demonstrated that Meta tracked and retained behavioral metadata, geolocation, and engagement telemetry from users under 13 without verifiable parental consent, subsequently using the datasets to train proprietary machine learning architectures and generative AI models.

The settlement arrives amidst broader multi-district litigation against social media platforms. Snap Inc., Alphabet's YouTube, and ByteDance's TikTok face thousands of coordinated state and personal injury lawsuits consolidated under US District Judge Yvonne Gonzalez Rogers in Oakland, California, addressing youth mental health crises. The settlement follows a series of state-level legal setbacks for Meta, including a $375 million jury verdict in New Mexico followed by an additional $567 million judicial penalty, alongside a joint $6 million damages award against Meta and Google by a Los Angeles jury.
//DBTech/IMS/MIS//

In 2009, Usain Bolt of Jamaica set a human world record of 9.58 seconds in the 100-meter dash. This past Saturday, in a five-day competition organized jointly by the Beijing Municipal Government and several organizations, a robot surpassed that record on August 26th. Earlier, in the preliminary rounds, one robot clocked 9.32 seconds and another, three days prior, finished in 9.39 seconds, already beating last year's record of 21.50 seconds. However, in the final stage, a Chinese robot shattered all records by completing the race in 8.86 seconds.

Upon crossing the finish line, the machine collided with a soft barrier and suffered a fall, emitting sparks around its waistline. Robotics specialists emphasized that post-sprint deceleration and dynamic stabilization present significant biomechanical hurdles.

Commenting on the engineering complexity, Professor Dennis Hong of the University of California, Los Angeles (UCLA) stated, "Engineering a bipedal robot to sprint at such velocity is an extraordinarily complex undertaking. As velocity escalates, sustaining equilibrium, absorbing mechanical impact, and controlling kinematic stability become progressively difficult, alongside managing actuator power and battery discharge rates."

Technical experts underscored that beyond sprint benchmarks, real-world deployment depends on operational safety, situational obstacle avoidance, and rapid recovery from balance disruptions. The event brings together over 2,000 robots from 666 teams across 16 countries including Japan, competing in over 1,300 categories such as football, long jump, and table tennis. Driven by state-level initiatives identifying robotics as a major strategic growth driver, commercial and research enterprises are rapidly accelerating technological frontiers in humanoid hardware.
//DBTech/NHK/SME/EMS//

According to the latest data from market research firm Counterpoint Research, overall smartphone sales in the region declined by 15 percent year on year due to rising component prices and inflation. Despite the downturn, Samsung was the only major brand to increase its shipments during the period.

Market Share and Leading Brands

Samsung remained at the top of the Southeast Asian smartphone market, commanding a 24 percent share. The company’s annual shipments increased by 6 percent, compared with the same period a year earlier.

Xiaomi ranked second with an 18 percent market share, although its sales remained largely stagnant compared with the previous year. Oppo slipped to third place with a 17 percent share, as its sales declined significantly from 22 percent a year earlier.

Transsion, the parent company of Infinix and Tecno, ranked fourth with a 15 percent share, while US technology giant Apple stood fifth with a 9 percent market share.

Declining Demand for Budget Phones, Premium Segment Surges

Counterpoint Research’s report highlighted a major shift in consumer demand. Sales of budget smartphones priced below $150 fell by as much as 38 percent compared with the previous year.

The market for mid-range smartphones priced between $250 and $499 also declined by 11 percent. The main reason was the rising cost of electronic components, which has made it increasingly difficult for manufacturers to maintain profitability on lower-priced devices, leading them to reduce supplies.

In contrast, sales of premium smartphones priced between $500 and $699 surged by a remarkable 74 percent. Sales of flagship smartphones priced above $700 also increased by 18 percent.

Outlook

Counterpoint Research has indicated that the pressure on the smartphone market could continue during the second half of the year, from July through December, following the trend observed in the first half.

With memory chip prices and inflation continuing to rise globally, smartphone manufacturers are unlikely to have much room to reduce prices. At the same time, consumers are becoming increasingly cautious about purchasing new phones. As a result, market analysts expect overall smartphone sales to decline somewhat further during the remainder of the year.

//DBTech/BMT/OR//

The Minister highlighted that drought and ecological degradation are intrinsically linked in Bangladesh. Across the northwestern region, nearly 5.46 million hectares of land remain drought-prone, where operating 16,000 deep tube-wells has led to critical groundwater depletion, prompting curbs on new irrigation projects. To mitigate these bottlenecks, the Bangladesh Agro-Meteorological Information Systems (BAMIS) is actively deployed to track drought metrics and transmit automated advisories to farmers via SMS, voice calls, and dedicated mobile applications.

He stated that the National Adaptation Plan (NAP) charts long-term technical mitigation priorities through 2050. Emphasizing basin-wide hydrological management across the Ganges-Brahmaputra-Meghna systems, he noted that dry-season flows are crucial for checking soil degradation. Domestically, under the National Green Mission, Bangladesh is planting 250 million trees, expanding coastal mangrove barriers against salinity ingress, and executing engineering plans to re-excavate 20,000 kilometers of canals and rivers over the next five years.

Welcoming the setup of the International Drought Resilience Observatory, the Minister proposed linking Bangladesh’s BAMIS infrastructure with the global monitoring network while urging advanced technology transfer to engineer salt-tolerant and drought-resilient crop varieties.

The high-level session was attended by foreign ministers, country delegates, and representatives from global agencies including FAO, ESCAP, IOM, UNICEF, UNDRR, UNDP, IFAD, AIIB, and the African Union.
//DBTech/THP/SME//

The memorandum of understanding was signed by Ukrainian President Volodymyr Zelenskyy and UK Prime Minister Andy Burnham during a bilateral meeting in Kyiv, the capital of Ukraine.

Battlefield Data and ‘Avengers Labs’

According to information published on the UK government’s website, the agreement will give British universities, researchers and technology companies access to the Ukrainian military’s specialised ‘Avengers AI Labs’.

According to Ukraine’s Ministry of Defence, the platform has been built around a massive dataset containing approximately five million labelled battlefield images. A large portion of the data has been collected through Ukraine’s own ‘Delta’ combat system.

Thousands of digital and infrared cameras deployed on the battlefield regularly contribute images of various targets, including tanks, artillery, air-defence systems and drones, to the database.

Technological Development and Automated Target Recognition

A report by The Kyiv Independent said AI models developed using this data have already enabled Ukrainian forces to operate an automated target-recognition system. The system analyses more than 100,000 hours of drone video footage every month and can immediately identify approximately 70 percent of enemy positions and targets.

Under the new agreement, engineers from the two countries will work on a specialised project to transform optical-fibre cables into AI-powered sensors that can be used to protect military bases.

The two countries will also conduct joint research into developing energy-efficient AI chips capable of operating with low power consumption for drones and autonomous systems.

First Foreign Trip and Missile Production Facility

This was Andy Burnham’s first foreign visit since taking office as UK prime minister. During the trip, he also co-chaired a meeting of the ‘Coalition of the Willing’.

According to a Reuters report, as part of the visit, Britain authorised missile manufacturer MBDA, which produces the long-range ‘SCALP’ cruise missile, to provide Ukraine with classified information related to UK-made components of the missile.

The decision will enable Ukraine to establish its own production and assembly line for the missile on Ukrainian territory.

//DBTech/BMT/OR//

The test was conducted by the channel’s lead presenter, Zack, who had tested the durability of Samsung’s ‘Galaxy Z Fold 8’ around two weeks earlier. While Samsung’s device successfully passed that test, Google’s latest foldable failed to impress.

Repeating the Same Design Flaw

As usual, the test began with a scratch test to assess the display’s resistance to marks, a flame test in which the screen was exposed to intense heat, and a test involving dust and sand to examine the hinge’s resistance to debris.

However, the major failure occurred during the ‘bend test’, in which the phone was forcefully bent backward by hand. Under pressure, the phone broke in the middle across the antenna line.

Analysts said Google has repeated a similar design flaw across four consecutive generations of its foldable smartphones. Because the body is structurally weaker around the area where the antenna is positioned, the devices have repeatedly broken at the same point when subjected to pressure.

Front Glass Comes Completely Loose

The latest test revealed not only a fracture across the antenna line but also another serious structural problem. Under extreme pressure, the front glass of the Pixel 11 Pro Fold completely separated from the main frame and came loose.

It is worth noting that last year’s ‘Pixel 10 Pro Fold’ was completely destroyed at the very end of its durability test after its battery was damaged and began emitting smoke. Although no similar fire-related incident occurred this time, the structural weakness of the new model was clearly exposed.

Repeated design flaws in a premium smartphone with a high price tag are putting Google on the back foot in the increasingly competitive foldable-phone market.

//DBTech/BMT/OR//

Under the proposed legislation, technology platforms that violate or fail to comply with the rules could face fines of up to 10 percent of their annual global revenue. The bill would also require technology companies to verify users’ ages accurately. For this purpose, platforms could use information from users’ existing accounts, facial-recognition technology or digital identification.

Tough Measures to Protect Children’s Mental Health

In a statement, Prime Minister Christopher Luxon said the government would not tolerate the toxic impact of social media on the new generation of children. He said harmful content, addictive technologies and various forms of social pressure on these platforms were seriously affecting children’s family lives, mental health, sleep and education.

Australia previously became the first country in the world to ban children under 16 from using platforms such as TikTok, YouTube, Instagram and Facebook. New Zealand is now following the neighbouring country’s lead with a similar strict policy.

Political and Implementation Challenges

However, there is considerable uncertainty over whether the government’s proposed bill will pass Parliament. New Zealand First, one of the key coalition partners in Prime Minister Luxon’s government, has already announced that it will not support the legislation.

The country’s Foreign Minister and leader of New Zealand First, Winston Peters, said in a post on the social media platform ‘X’ that such legal restrictions could have a dangerous impact on the country’s tradition of individual freedom. He believes that responsibility for keeping children away from social media should primarily rest with parents rather than the government.

Despite political opposition and technical challenges, Prime Minister Luxon said that although it may not be possible to make the legislation 100 percent perfect, it is essential to make an effort to protect children.

//DBTech/BMT/OR//

The move signals Xiaomi’s ambition to further consolidate its position alongside global rivals such as Apple, Samsung and Huawei.

The new chip will serve as the successor to the company’s first in-house processor, the ‘XRing O1’, which entered the market last year.

Production Using TSMC Technology and Target

According to several sources familiar with the matter, chip manufacturer TSMC will produce Xiaomi’s new processor using its advanced 3-nanometre manufacturing technology. The new ‘XRing O3’ processor is reportedly planned for use in Xiaomi’s upcoming flagship foldable smartphone.

Initially, Xiaomi is targeting the release of between 200,000 and 300,000 smartphones powered by the new processor. The initiative is primarily aimed at challenging the dominance of Chinese rival Huawei in the premium foldable smartphone market.

Market Challenges and Future Plans

Rising prices of memory and various components have increased smartphone production costs worldwide, contributing to a slowdown in the global mobile market. In this challenging environment, Xiaomi considers reducing its dependence on chip suppliers such as Qualcomm and MediaTek to be particularly important for commercial security.

In addition to its main mobile processor, Xiaomi has also commissioned TSMC to manufacture the 6-nanometre ‘XRing O100’ chip for running its artificial intelligence ‘MiMo’ model and the 3-nanometre ‘XRing D100’ chip for autonomous-driving technology.

The new chips are expected to be ready for final deployment by next year.

//DBTech/BMT/OR//

According to a report by US media outlet Semafor, Bilibili has relaunched its app on Android to make it more accessible to international users, while an iOS version is expected to arrive soon. At the same time, the company is developing a full English-language version of its website to make it easier for creators from other countries to upload videos.

Relaxed Rules and New Features

One of the biggest changes in Bilibili’s push into international markets concerns account verification or identity verification. Until now, users outside China who wanted to post videos were required to submit a passport or another government-issued identification document.

The new international app has completely removed this requirement. From now on, content from inside and outside China will be displayed together on the app.

To connect content creators and audiences from around the world, Bilibili is also appointing regional representatives in cities including Los Angeles, London, Mexico City, São Paulo, Istanbul and Tokyo.

Political Pressure and the Challenge of Global Competition

Bilibili’s rapid global expansion has sparked fresh discussion in the technology market. Chinese video platform TikTok previously faced intense scrutiny in the United States over national security concerns, followed by legal pressure to sell its US business. This has raised questions about whether Bilibili could face a similar situation if it becomes sufficiently popular in international markets.

However, if Bilibili succeeds in attracting international YouTubers amid competition from YouTube and Netflix in the video and podcast space, the move could bring significant changes to the global video-sharing market.

//DBTech/BMT/OR//

Although Samsung has not revealed the name of any specific device for the event, technology insiders strongly believe that the company could unveil the new ‘Samsung Galaxy S26 FE’ (Fan Edition). Samsung typically introduces its FE models to offer fans a more affordable smartphone while retaining the key attractions and flagship experience of its premium models.

Smartphone Features and Technology

In a statement, Samsung said the new Galaxy device will provide consumers with an impressive experience based on the core S26 series. The device is expected to feature powerful camera capabilities, Samsung’s artificial intelligence technology ‘Galaxy AI’ and the company’s updated operating system, ‘One UI’.

As a result, users will be able to enjoy premium features and capabilities at a comparatively lower cost.

Expected Specifications and Camera

According to information from various technology-focused leaks, the Galaxy S26 FE could feature a 6.7-inch AMOLED display with Full HD+ resolution and a 120Hz refresh rate. The smartphone may be powered by an ‘Exynos 2500’ chipset.

For photography, the phone is expected to feature a triple-camera setup on the rear, led by a 50-megapixel main sensor.

The camera system is also expected to include a 12-megapixel ultrawide lens and an 8-megapixel telephoto lens.

The event will be streamed live on Samsung’s official website and YouTube channel on August 27. Technology enthusiasts will be able to join the event online to learn more about the device’s final design, memory variants and pricing.

//DBTech/BMT/OR//

The company said the outage, which lasted more than seven hours last Monday, was not caused by any malicious code or configuration change. Instead, the disruption resulted from an unprecedented surge in traffic.

The network failure occurred because the core infrastructure of GitHub’s US-based data centre was unable to automatically adapt to the additional load.

The Surge of AI-Generated Code Behind the Outage

An internal GitHub report found that the number of ‘commits’, or changes made to code files, more than doubled in just four months, from April to mid-August. While the platform recorded 1.4 billion monthly commits in April, the figure jumped to 2.9 billion four months later.

According to technology experts, the surge is directly linked to the dramatic increase in the use of artificial intelligence to generate programming code. The rapidly growing global adoption of AI coding tools from technology companies such as OpenAI and Anthropic has placed enormous pressure on GitHub.

Infrastructure Overhaul and Future Challenges

GitHub acknowledged that the rapid growth in the number of users helps explain the increasing pressure on its systems, but said this cannot serve as an excuse for service outages.

To prevent similar network disruptions in the future, GitHub is reviewing its server processing capacity and memory-alert mechanisms. The company is also working to redefine retry limits and timeouts to prevent excessive simultaneous pressure on the network when information is exchanged between multiple servers.

Despite these technical measures, technology analysts believe they may not be enough to completely eliminate the long-term pressure on GitHub. Dependence on AI technology in modern software development continues to grow rapidly.

As the primary platform for professional and hobbyist coders around the world, GitHub will therefore have to continually adapt to the technological changes and increasing workload generated by AI-assisted development.

//DBTech/BMT//

The settlement represents one of the largest financial payments in the history of enforcement actions involving the Children’s Online Privacy Protection Act, or ‘COPPA’.

Background of the Case and Main Allegations

In 2024, during the administration of then-US President Joe Biden, the US Department of Justice took legal action against TikTok. The allegations stated that TikTok and its predecessor app, Musical.ly, allowed children under the age of 13 to create accounts without obtaining permission from their parents.

The companies were also accused of illegally collecting children’s personal information, including email addresses, from users of the ‘Kids Mode’ feature.

In addition, although parents requested that their children’s accounts and data be deleted, the company repeatedly failed to comply with those requests, which constituted a clear violation of US technology laws.

Structural Changes and Terms of the Agreement

The US Department of Justice said in a statement that TikTok’s ownership structure, management and user-safety policies have undergone major changes since the case was filed.

A joint entity has been established with investments from major companies including Oracle, Silver Lake and MGX to operate TikTok’s US business, while ByteDance retains a 19.9 percent stake. Under the new structure, age restrictions for younger users and parental oversight mechanisms have been significantly strengthened.

Under the settlement agreement, TikTok will immediately pay $300 million. The remaining $100 million will be paid after a court formally vacates a previously issued legal order involving Musical.ly.

With the payment, TikTok has resolved a major long-running legal challenge related to its handling of children’s privacy.

//DBTech/BMT/OR//

Buying a first camera used to involve considerable calculation for newcomers. They had to choose a particular brand, set a budget and decide how much zoom they needed before making a purchase. Today, however, the question at the very top of the checklist is: DSLR or mirrorless?

For a new user looking to buy a camera today, the simple answer is mirrorless. Over the past few years, leading camera manufacturers have introduced most of their latest autofocus technologies, video capabilities and new lenses for their mirrorless product lines. From APS-C sensor models designed for beginners to full-frame models for professionals, mirrorless cameras have become the most attractive choice for users at virtually every level.

That does not mean DSLR cameras have suddenly become inferior. The main point is that the camera industry’s technological direction is no longer toward DSLRs. Both DSLR, or digital single-lens reflex, and mirrorless cameras are digital cameras. The fundamental difference, however, lies in how light coming through the lens reaches the user’s eye and the camera’s sensor.

What Changed When the Mirror Was Removed?

Inside a DSLR camera, there is a ‘mirror’ made of glass that reflects light coming through the lens toward the optical viewfinder. When the shutter button is pressed, the mirror flips upward and the sensor captures the image. In a mirrorless camera, this mechanical mirror system does not exist. Light falls directly onto the sensor, and the live feed from the sensor appears on the rear display or electronic viewfinder.

Technically, this may appear to be a small change, but it has brought about a major revolution in the way cameras are used.

Because the sensor in a mirrorless camera is directly exposed to the scene, the camera can offer many more autofocus points. It can easily detect human eyes, animals and vehicles. Users can even see an accurate preview of the image’s exposure on the screen before taking the photograph. This makes it considerably easier for new photographers to capture the desired image.

Moreover, because there is no mechanical mirror that needs to move up and down, mirrorless cameras can use electronic shutters to capture a large number of images at extremely high speeds.

DSLRs, however, have some advantages of their own. Looking through their optical viewfinders does not require additional power from a digital display, resulting in generally better battery life than mirrorless cameras. On the other hand, although mirrorless cameras are smaller and lighter, that advantage can quickly disappear when a large telephoto lens is attached to the front.

In terms of image quality, there is no particularly obvious difference between the two technologies. With a good sensor combined with a high-quality lens, both types of cameras can produce outstanding photographs. However, because of their faster autofocus, higher shooting speeds, advanced video capabilities and overall ease of use, major companies such as Canon and Nikon have shifted their focus toward mirrorless systems.

Is There Still a Reason to Use a DSLR?

For someone buying an interchangeable-lens camera from scratch, the decision is fairly straightforward: choose a good mirrorless camera.

For those who already use a DSLR, however, the calculation is different. If your camera bag contains a large collection of expensive Canon ‘EF’ or Nikon ‘F’ lenses and other accessories, abruptly replacing the entire system with newer technology can be costly and unnecessary. Moreover, lens adapters make it possible to use older DSLR lenses with new mirrorless camera bodies.

Even so, the recent trend among professional photographers is clearly toward mirrorless cameras. A survey of 1,000 photographers conducted by the photo blog ‘Shotkit’ found that 63 percent of the professionals surveyed now use mirrorless cameras.

The market transformation is also evident from the product lines of Canon and Nikon. Nikon has not introduced a new DSLR since 2020 and has instead focused entirely on its ‘Z’ series mirrorless system. The notable exception is ‘Pentax’; its parent company, Ricoh, has said it will continue working with SLR technology.

According to statistics, approximately 90 percent of interchangeable-lens cameras shipped worldwide in 2025 were mirrorless. Overall, DSLRs remain capable cameras, are familiar to and used by millions of photographers, and can be an excellent option for buyers seeking affordable equipment in the second-hand market. However, at this point, building an entirely new DSLR-based setup or system from scratch does not make much technological sense.

//DBTech/BMT/OR//

Although passkeys are considerably more advanced than the traditional system, it is important to understand some of their specific features and how they work. They do not rely on information that users have to remember; rather, they are linked to a specific device or software. Therefore, users need to be careful not to lose access to the device or software where their passkeys are stored. Unlike passwords, passkeys also cannot be easily shared with friends.

However, for those who are not yet using a ‘password manager’ to protect their accounts, switching from conventional passwords to passkeys is arguably the most convenient and secure option.

Once users become familiar with how passkeys work, they can significantly reduce the hassle of everyday logins. The most important consideration is how and where the passkeys are stored. The good news is that almost all popular operating systems and password managers now offer passkey storage. Here is a closer look at the technology.

Technical Difference Between Passwords and Passkeys

• Password: A specific text or secret code. Websites encrypt these passwords and store them on their servers in the form of a ‘hash’. During login, the encrypted representation of the entered password is compared with the information stored on the server before access is granted.

• Passkey: A passkey does not require users to remember any text or code. Instead, it works using two cryptographic keys: a ‘public key’ and a ‘private key’. The public key is stored on the website or platform’s server, while the private key is encrypted and securely stored on the user’s own phone or computer. During login, the key is activated after verification through the device’s Face ID, fingerprint or PIN.

Password Problems Solved by Passkeys

• Phishing protection: A passkey is digitally and permanently associated with a specific domain or website. As a result, even if users accidentally open a fake or fraudulent website, the passkey will not disclose any information there.

• No risk of weak or reused passwords: Users often create weak passwords because they are easier to remember or use the same password across multiple websites. Since passkeys use cryptographic codes, they are inherently strong and cannot be reused across different services.

• Protection against data breaches: Even if hackers manage to breach a website’s database, they can only obtain the publicly available ‘public key’, which cannot be used to log into an account. The crucial ‘private key’ remains securely stored on the user’s own device.

• Automatic two-factor security (2FA): With a passkey, possession of the physical device and biometric or PIN verification take place together. This reduces the need for separate two-factor authentication in many cases.

Challenges of Using Passkeys and What to Do

• Ecosystem limitations: Apple Passwords, Google Password Manager and Windows all have their own systems for storing passkeys. However, if you use devices from different brands, such as an Android phone and a MacBook, using a reliable third-party ‘password manager’ or a physical security key, such as a YubiKey, may be the most convenient option.

• Difficult to share: Unlike an ordinary password, a passkey cannot simply be shared with someone verbally or in writing. Sharing requires a shared vault provided by a compatible password manager.

• Current availability: Passkeys are still not available on every website. However, on websites that support passkeys, enabling the feature through their security settings is a safer option. For services that do not yet support passkeys, using a strong password through a password manager remains the preferred approach.

//DBTech/BMT/OR//

Today, pen drives from brands such as Kingston generally start with a minimum capacity of 64GB and go up to 512GB or even higher. As a result of technological advancement, small-capacity memory drives have gradually lost their commercial relevance.

The major factor behind the dramatic increase in flash drive capacity has been the development of ‘NAND’ technology inside memory chips. Over time, researchers have reduced the size of cells inside NAND chips and, through three-dimensional or ‘3D NAND’ technology, have learned to stack cells on top of one another. This has created opportunities to store many times more data without increasing the physical size of pen drives.

When Toshiba and SanDisk unveiled the first 8-gigabit NAND chip in 2005, a new era of storing twice as much data at lower costs began. Subsequently, the development of 3D and TLC NAND technologies led to a rapid decline in storage prices.

The Economics of Not Making Small Drives

Manufacturing a 1GB pen drive has not become technically impossible; rather, selling one has become commercially unprofitable. Producing a pen drive requires not only a memory chip but also a USB connector, circuit board, plastic or metal casing, controller and packaging. Even if the capacity of a 64GB pen drive is reduced to 1GB, the costs of auxiliary components, transportation and marketing remain almost the same.

As a result, there is little difference in the final production cost of 1GB and 64GB pen drives. Moreover, modern computer operating systems or Windows installations themselves require at least 8GB of storage, making small-capacity drives impractical for everyday use.

Special Uses in the Industrial Sector

Manufacturers are gradually reducing the supply of older-generation NAND chips. In 2026, global MLC NAND chip production is forecast to decline by approximately 41.7 percent.

However, although they have become obsolete for general consumers, small-capacity flash drives of 256MB or 1GB are still used in various specialized types of industrial equipment. In such cases, consistent reliability of the equipment and compatibility with specialized software are more important than having enormous storage capacity.

//DBTech/BMT/OR//

The service normally costs $200 a year, or $20 a month, for regular customers. The move is believed to be aimed at increasing the number of users and engagement with Google’s chatbot, Gemini, by offering students special access to the service.

Benefits for Students

Under the ‘Google AI Pro’ subscription, students will receive four times more access to Google’s artificial intelligence model ‘Gemini’ than regular users. They will also have access to ‘Spark’, an automated agent-based platform. In addition, Gemini will be directly integrated into Google Workspace applications such as Gmail and Docs.

The offer is not limited to artificial intelligence features. Students who are not particularly interested in using AI will also find significant benefits in the package. Under the promotion, users will receive 5 terabytes of cloud storage free of charge. The package also includes ‘Google Health Premium’, which can be connected to Fitbit and other health-tracking devices.

Other Offers and Important Cautions

Google has also introduced an opportunity for students outside the United States. Students in other countries will receive a one-year free subscription to ‘Google AI Plus’. It is essentially a basic service priced at $5 per month, offering access to ‘Gemini Omni’, double usage limits and 400GB of cloud storage.

In addition, Google has introduced an affordable bundle combining AI Pro and YouTube Premium, which will be available for just $9 per month.

However, students should be particularly careful about one issue when taking advantage of the offer. Once the specified one-year period ends, the service will automatically renew and the regular subscription fee will begin to be charged.

Therefore, it is important to remember to cancel the subscription before the free one-year period expires. This is particularly significant for students who plan to make extensive use of the cloud storage.

//DBTech/BMT/OR//

For now, Android users in the United States, Canada, the United Kingdom, Ireland, Australia, New Zealand and Sweden can access the feature directly from Spotify’s ‘Fitness Hub’.

Customization and Beat Matching Features

The new feature offers users 25 different presets based on various workout and music categories. Users can set the duration of their run and activate the ‘Beat Matching’ feature to maintain a specific pace.

The Beat Matching feature synchronizes the rhythm or beat of songs with the user’s running pace. In addition, users can choose whether to turn on or off necessary audio instructions, or ‘audio cues’.

The Fusion of AI and Fitness

Spotify has enhanced the new Running Mode based on an initial experiment launched in 2018 and its AI-powered ‘Prompted Playlists’ feature, introduced earlier in 2026.

The technology company said one of the most popular prompts among users on its platform was finding music for fitness or exercise. Taking this strong demand into consideration, Spotify has integrated AI technology with music streaming and workout features.

//DBTech/BMT/OR//

What Is Changing?

Previously, for regular or long-form YouTube videos, a view generally required at least 30 seconds of viewing before it was counted. Last year, YouTube introduced immediate view counting for Shorts, where a view is registered as soon as playback starts.

Now, the same system will be extended to all types of long-form videos and live streams, alongside Shorts. The move is primarily intended to bring YouTube’s view-counting system more closely in line with those of competing platforms such as TikTok and Instagram.

Why the Change and What Will It Mean?

YouTube said in a blog post that having different view-counting systems for different types of videos often confused content creators. Creators had long been asking for a unified method of measuring views so they could more accurately assess the actual reach and impact of their content.

Under the new system, the publicly displayed view count will increase more quickly than before.

However, the change will not affect creators’ earnings or YouTube’s advertising policies. YouTube has clarified that creator revenue will continue to be determined based on factors such as how long viewers actually engage with and watch the content.

For more detailed performance analysis, YouTube will continue to provide “Engaged Views” in the Advanced Mode of its Analytics dashboard. This will allow creators to distinguish between a simple video start and more meaningful viewer engagement.

//DBTech/BMT/OR//

The company has already informed its suppliers that, after withdrawing production from China, its smartphones, smartwatches and wireless earbuds will primarily be manufactured in Vietnam and India.

Vietnam’s Success and Global Manufacturing Shift

Earlier this year, reports indicated that Google had already begun producing its new Pixel 11 series smartphones in Vietnam.

The company has successfully completed the complex process of building manufacturing infrastructure, conducting testing and establishing quality control systems necessary for producing flagship smartphones in a new country.

Having achieved success in manufacturing advanced devices such as smartphones in Vietnam, Google now finds it easier to expand production there to accessories such as earbuds and smartwatches.

Technology analysts note that Google's transition away from China is easier than that of US rival Apple. One major reason is that Google does not officially sell Pixel phones in China, and its global customer base is significantly smaller than Apple's.

In addition, Google can benefit from the extensive supply-chain ecosystem already built in Vietnam by competitor Samsung.

Production Targets and AI Strategy

Google has informed suppliers that it aims to increase Pixel phone shipments by 8 to 10 percent in 2026, compared with the 12 million units produced in 2025.

A key objective behind expanding Pixel sales is to bring its in-house artificial intelligence service, Gemini, to a larger number of users.

Meanwhile, to cope with rising global memory-chip prices, Google has adopted a strategic approach by combining purchase orders for memory chips used in smartphones with those used in its cloud-computing business. This allows the company to negotiate more efficiently and better control production costs.

//DBTech/BMT/OR//

However, not all of CapCut's AI features are necessarily available for free.

Another useful option is InVideo AI, particularly for turning a written script directly into a video without requiring any pre-recorded footage. Users can enter a topic or text, and the app automatically arranges scenes, voiceovers and subtitles to create an initial video.

However, its free version comes with limited usage and places a watermark on exported videos.

Special Features for YouTubers

For YouTube content creators, YouTube Create is a particularly useful alternative. It is especially helpful for creators who record videos on mobile phones, with features designed to improve audio quality and synchronize video with music.

Using Google's Gemini-powered AI technology, users can also create short video clips with simple prompts or perform basic editing using their preferred photos and videos.

One of its biggest advantages is that creators can publish their finished videos directly to their YouTube channels from within the app.

A Platform for Professional and Long-Form Projects

For complex, long-term projects beyond quick social-media videos, DaVinci Resolve is one of the most reliable options. Designed primarily for computers, the software combines color correction, sound editing and a wide range of visual effects, making it possible to edit an entire feature film, drama or documentary.

However, its interface is considerably more complex than those of typical mobile apps, meaning that beginners may need more time to learn how to use it effectively.

In addition, many of DaVinci Resolve's most advanced AI features are available only in the paid Studio version.

Therefore, creators should choose their editing platform based on the type of content they produce and the features available in each tool's free version.

//DBTech/BMT/OR//

The two companies have already reached a formal agreement under which 100% of Lingshi Games' shares will be transferred. Reuters confirmed the development in a report.

In an internal memo sent to Lingxi Games employees, the company's CEO Zhou Bingshu said that despite the change in ownership, the existing management team under his leadership would continue to run the game studio.

The memo did not specify the final value of the transaction or when the transfer would be completed. However, a separate Bloomberg report valued the deal at at least $1.5 billion.

Greater Focus on AI and Cloud Technology

As part of its strategy to strengthen its core businesses, Alibaba has recently been selling assets and companies outside its main areas of operation.

The company is now focusing more heavily on emerging sectors such as artificial intelligence (AI) and cloud computing, increasing investment and resources in these areas.

Earlier, Alibaba sold its ownership of the hypermarket chain Sun Art and department-store operator Intime. The decision to sell its gaming unit is part of the same broader strategy.

A New Chapter for Lingxi Games

Guangzhou-based Lingxi Games is best known internationally for its strategy game Three Kingdoms: Strategy Edition, which was developed in collaboration with renowned Japanese game developer Koei Tecmo.

The company's fundraising plans had stalled in 2023 amid tighter Chinese government restrictions on the gaming industry. However, the gaming market has since begun to recover.

Company officials expect the involvement of TruStar Capital to give the studio an opportunity to expand its business and begin a new phase of growth.

//DBTech/BMT/OR//

On Monday, Germany's Federal Cartel Office officially announced the conclusion of its year-long investigation and outlined the changes Apple must make.

Competition Law and the New Policy

The German regulator's investigation found that Apple's App Tracking Transparency (ATT) system was giving Apple's own apps an advantage over third-party applications.

When third-party apps sought users' permission to access and use their data, the pop-up messages presented to users contained certain wording or symbols that could discourage them from giving consent. The authority concluded that this created an uneven competitive environment in the market.

Under the new decision, Apple will have four months from receiving the notice to implement the proposed changes. An independent trustee will then monitor Apple's proper implementation of the measures for the following seven years.

Impact on Advertising and the European Union

Under the new rules, pop-up messages used by third-party apps to request users' consent must be completely neutral in terms of both their visual presentation and wording.

This will give major technology companies such as Meta, the parent company of Facebook, as well as other app developers greater flexibility in obtaining users' consent.

Targeted advertisements based on reliable user data can generate significantly more revenue than conventional advertising because they allow advertisers to reach specific groups of consumers more precisely.

Apple said the new policy will be implemented in almost all European Union countries.

Earlier, France and Italy had also fined Apple over the same controversial tracking system, imposing penalties of €150 million and €98.6 million, respectively.

//DBTech/BMT/OR//

The Trump-backed crypto company World Liberty Financial is working in partnership with Hong Kong-based platform WorldXL, which provides users with access to AI models developed by several Chinese companies that the US government has identified as potential national-security risks.

The information emerged from an investigation by international news agency Reuters.

Chinese AI Models and Financial Gains for the Trump Family

Established this year, the WorldXL platform provides customers with access to a total of 90 AI models. According to the Reuters investigation, 43 of those models were developed by Chinese technology companies such as Alibaba, Baidu and Z.AI.

The US Departments of Defense and Commerce have identified some of these companies as having links to the Chinese military or involvement in the theft of US technology.

Yet the platform accepts USD1, the cryptocurrency issued by Trump's company World Liberty Financial, as a means of paying for access to its AI services. Since the Trump family owns a 38% stake in World Liberty, a portion of the profits generated from each transaction ultimately goes into the Trump family's financial interests.

President Trump's two sons, Donald Trump Jr. and Eric Trump, have also promoted the joint service on social media.

Legal Status and Policy Contradiction

There is currently no direct legal restriction preventing ordinary US users from accessing these Chinese AI models. However, technology and political analysts argue that the arrangement raises serious questions about conflicting interests.

The Trump administration has taken a tough stance on limiting the spread of Chinese technology. Against that backdrop, they argue, generating revenue through Chinese technology via a company owned by Trump's family amounts to a policy contradiction and raises ethical concerns.

However, both the White House and World Liberty have denied allegations of a conflict of interest.

Their representatives argue that such technology transactions are a normal part of the global market and that the arrangement does not violate US policy.

//DBTech/BMT/OR//

A key part of Android's security is Google Play Protect. It scans apps when they are downloaded from the Google Play Store and continues to regularly scan the handset after installation to detect potentially harmful software or malware. Android's built-in sandboxing technology also prevents one app from accessing another app's data without authorization.

Regular security updates and permission-based access have made modern smartphones considerably more resistant to cyberattacks. Technologies using artificial intelligence are also being developed to detect and prevent fraudulent calls and malicious messages.

New Cyber Risks and Their Limitations

Another major reason for the declining need for antivirus apps is that the nature of today's primary cyber threats has changed. Rather than traditional computer viruses, social engineering and psychological cyber fraud have become major risks.

Users are increasingly targeted through enticing fake messages, password-stealing scams, fake help desks, calls from people posing as bank representatives, and malicious website push notifications.

If a user inadvertently provides personal information through a fraudulent link, however, even an antivirus application may not be able to protect them.

When Might an Antivirus App Be Necessary?

There are certain situations in which an additional antivirus app can provide useful protection. Users who sideload APK files from unfamiliar sources outside the Google Play Store face a greater risk of introducing malware to their phones.

Extra protection may also be reasonable for older smartphones that no longer receive regular security updates, as well as devices without Google Play Services. In such cases, installing a reputable antivirus application can provide an additional layer of security.

Ultimately, there is no reason for most Android users to be worried simply because their phones do not have an antivirus app.

Rather than relying solely on software, users should avoid clicking unfamiliar links, exercise caution when using public Wi-Fi, and develop safe internet habits. In today's technology-driven world, such awareness remains one of the most effective ways to stay protected.

//DBTech/BMT/OR//

The latest Android 17 build introduces an integrated layout editor, allowing users to customize almost all aspects of the Quick Settings panel according to their preferences.

With the new layout editor, users can choose which features should appear as larger tiles. They can also reposition the brightness slider and media player—used for controlling music and video playback—by dragging them higher or lower on the screen.

As a result, Pixel users will have greater freedom to customize the Quick Settings panel instead of being restricted to Google's predetermined layout.

More Customization and Visual Changes

Google is also planning to give users greater control over theme selection in the next major version of Android.

The new interface will include a dedicated slider that allows users to adjust the hue of four primary colors used across their device according to their preferences.

In addition, Google is working on adding new blur effects to various visual elements of the interface to enhance its overall appearance.

Enhanced Security to Combat Social Engineering

The update also places greater emphasis on protecting users from social engineering attacks, in which scammers manipulate people through fear, deception or enticing offers.

A new Call Protection feature has been introduced for this purpose. Under the new system, whenever a user attempts to dial a call-forwarding code, a confirmation dialog will appear on the screen.

This additional confirmation step is designed to make it more difficult for scammers to secretly redirect a user's calls to another number by exploiting the user's confusion or lack of awareness.

//DBTech/BMT/OR//

According to a report by international news agency Reuters, the model was trained with assistance from Qwen, the AI technology developed by Chinese e-commerce and technology giant Alibaba.

While Apple relies on third-party AI models developed by companies such as Google, OpenAI and Anthropic in other markets, the company has made a major strategic shift in its AI approach to strengthen its position in the Chinese market.

Major Achievement as First Foreign Company

Foreign technology companies often face various regulatory hurdles when seeking approval to operate in China. Against this backdrop, Apple is set to become the first foreign technology company to receive the Chinese government's green light to launch a customized AI model using its own technology.

The new Apple Intelligence features are expected to become available to iPhone, iPad, Mac and Vision Pro users in China through software updates over the coming months.

Apple has also recently published a guide explaining how Mac users can integrate Alibaba's Qwen model with Siri.

A Move Tailored to Local Regulations

Because of China's local data-protection laws and strict censorship policies, US technology giants cannot simply deploy the AI models commonly used in other parts of the world directly in the Chinese market.

To address this challenge, Apple has worked with local technology partner Alibaba to develop the model in a way that complies with China's legal and regulatory framework.

The addition of these new AI features in China could be particularly important for Apple's business at a time when smartphone sales have slowed somewhat in the global market. Given China's enormous consumer market, strengthening Apple's AI offerings there could help the company maintain its competitiveness.

//DBTech/BMT/OR//

According to a report by The Wall Street Journal, the two parties offered $60.50 per PayPal share in July, valuing the company at approximately $53 billion. PayPal rejected the proposal at the time, considering it insufficient.

Before the offer was made, PayPal's share price had fallen to one of its lowest levels in recent memory. The company's market value reached approximately $320 billion at the height of the pandemic-era boom but later declined to around $40 billion.

Despite the rejection of the initial proposal, negotiations between the parties have continued. The two sides are now reportedly discussing a new deal involving a higher share price. If negotiations progress smoothly, an official announcement of the potential mega-deal could come within the next few weeks.

Joint Ownership and Control of a Huge Market

According to international news agency Reuters, if the deal is completed, Stripe and Advent International would each receive an equal ownership stake in PayPal.

There are reportedly no plans to break up or separate PayPal. Instead, the company would continue operating under joint ownership.

If completed, the acquisition would make Stripe one of the world's largest online payment providers. Together, the combined operation is expected to handle approximately $3.7 trillion in transactions annually.

Enrique Lores took over as PayPal's chief executive officer in March this year as part of an effort to reorganize the business. He divided the company into three major divisions: Checkout, Venmo, and Payments & Crypto.

It remains unclear, however, how the potential acquisition would affect PayPal's internal structure.

Competition and Strategic Advantages

Technology analysts believe a potential combination of Stripe and PayPal could create a significant new alignment in the global payments industry.

The deal could substantially reduce Stripe's dependence on Visa and Mastercard for payment infrastructure.

At the same time, Stripe would gain direct access to several of PayPal's established services, including its popular Venmo payment platform, its well-established Checkout system and its cryptocurrency-related capabilities.

Such a combination could give Stripe a much broader presence across online payments while bringing together two major players with complementary strengths.

For now, however, it remains uncertain whether the ongoing negotiations will ultimately produce a successful agreement or collapse.

//DBTech/BMT/OR//

In an announcement on Friday, Google said that although users will be able to remove visible watermarks, the underlying digital identification and tracking mechanisms embedded in the content will remain fully intact.

Josh Woodward, vice president of Google's Gemini division, said in a post on X that the feature will be available for content generated by Google's Nano Banana, Omni and Lyria models.

Users will be able to turn visible watermarks on or off through the Gemini app and Google's video-editing platform Flow. The feature is also expected to be integrated into Google's search infrastructure soon.

Change Aimed at Professional Use

Visible watermarks can sometimes reduce the usability and visual appeal of content intended for creative or professional work. At the same time, being able to identify whether content was generated by AI has become increasingly important for digital safety and transparency.

Woodward said Google is therefore attempting to strike a balance between users' creative freedom and AI safety.

Although visible watermarks will become optional, the SynthID watermark embedded within Gemini-generated content and C2PA metadata, an international standard for establishing the provenance of digital content, will remain unchanged.

As a result, even when the visible watermark has been removed, users will still be able to verify through Gemini or Google Search whether an image or video was generated using AI.

Watermark Can Be Easily Disabled

Google is expected to make the feature available to users within the next few days.

Users will be able to enable or disable the visible watermark by going to Settings in the app and selecting Media Watermark.

Google has also released a new open-source library called Credentialo, designed to allow developers to incorporate local content-verification capabilities into their own applications.

Google Takes a More Flexible Approach

Recently, technology company Anthropic decided to add watermarks to text and files generated by its Claude models as part of efforts to comply with European Union AI regulations.

Following that controversial move, Google has introduced a more flexible approach: users will have the freedom to remove the visible watermark, while the underlying digital protection and identification mechanisms will remain active.

//DBTech/BMT/OR//

Under Apple's proposal, apps that normally pay a 30% commission on purchases made within the App Store would be subject to a 15% commission when transactions are made through external payment links.

Apple has proposed a 10% commission for its News Partner, Video Partner and Mini Apps Partner programs, as well as for subscription renewals.

Lower Commission for Small Developers

For developers participating in Apple's Small Business Program, the proposed commission would be just 5%.

Apple said that the vast majority of developers using its platform qualify for the small-business concession.

Apple and Epic's Legal Battle

The legal dispute between Apple and Epic Games began in 2020. In 2021, U.S. District Judge Yvonne Gonzalez Rogers ordered Apple to allow developers to provide external or third-party payment links within their apps, enabling them to avoid Apple's mandatory 30% commission.

Apple subsequently continued to seek fees ranging from 12% to 27%, prompting Epic Games to accuse the company of violating the court's order.

An appeals court later upheld the judge's earlier ruling but overturned the requirement preventing Apple from charging any fee on external payments. The court must therefore determine what level of commission would be reasonable for external transactions.

Epic Games has rejected Apple's latest proposal.

Court Also Tightens Rules on Google

Google is simultaneously engaged in a separate legal battle with Epic Games.

In June, Google introduced an external billing system on the Play Store and reduced its commission to 10%. In July, it also allowed third-party or alternative app stores to be used on the Android platform.

Epic Games subsequently complained to the court that Google had made the process of installing another app store on Android unnecessarily complicated.

The court took the complaint into consideration and ordered Google to remove the barriers and restrictions that make it difficult to install alternative app stores.

//DBTechBMT/OR//

Farmers in Gangwon, a mountainous province in northeastern South Korea, told local media that potato fields that appeared normal just a week earlier are now showing widespread crop damage when farmers attempt to harvest them.

Lee Sang-hyeok, a potato farmer in Hyeon, Yanggu County, near the North Korean border, told local broadcaster SBS that he had never witnessed such conditions in 40 years of farming. He said, ‘This isn't just an isolated incident involving one or two plots of land. I'm at a loss for words to express how I feel; I'm deeply hurt.’

The extreme heat has damaged not only potatoes but also cabbage grown across several thousand pyeong of farmland in Gangwon. Farmers had delayed harvesting in the hope of receiving better market prices. The intense heat, however, caused the cabbage to burn and rot in the fields, resulting in losses reportedly worth tens of millions of Korean won.

Various fruits have also been affected. Excessive sunlight has caused ‘sunburn’ on apple skins, with exposed areas becoming scorched and discolored. Peaches, meanwhile, are splitting because of the extreme heat.

Apple farmer Chon Jong-tae said apples expected to be ready for harvest around the Chuseok holiday at the end of September have stopped growing normally. Only the parts facing the sun are turning red, and if the heatwave continues, even those areas could become discolored.

Peach farmer Park Won-sun said farmers have been forced to return to their fields despite government advice to avoid outdoor work during the daytime heat. In his words, the peaches are rotting, leaving them no option to remain idle.

The heatwave is also affecting people. According to the Korea Disease Control and Prevention Agency, at least 31 people have died so far from the heatwave. More than 900,000 livestock and nearly 1.4 million farmed fish have also died during the same period.

On August 2, temperatures in the southeastern city of Yangsan reached 42.5 degrees Celsius. This is the highest temperature recorded in South Korea since modern temperature records began in 1904.

Farmers themselves are suffering physically from the extreme heat. At the same time, the threat of crop losses makes it difficult for them to follow government advice to remain indoors during the hottest part of the day.

‘This weather is becoming the new normal’

South Korean President Lee Jae-myung told cabinet members that people in the country had never experienced weather conditions like these before. He called for major reforms to the national crisis-response system.

According to the president, such extreme weather is gradually becoming the new normal.

So far, official damage assessments have focused mainly on human deaths and livestock losses. The total damage to agriculture has yet to be determined.

Farmers fear that widespread crop losses could push already high food prices even higher, particularly ahead of Chuseok, one of South Korea’s major autumn holidays.

Potatoes vulnerable to extreme heat

Researchers have previously identified potatoes as a crop particularly vulnerable to extreme heat. A study published in Potato Research in 2024 said South Korea’s summer potato production may require heat-tolerant varieties under future climate conditions. Planting spring potatoes relatively earlier could also provide some benefits.

South Korea’s experience is another example of increasingly severe heatwaves across Asia. Temperatures have reached 40 degrees Celsius or higher in several parts of Japan. Hong Kong recorded 36.9 degrees Celsius, its highest temperature since records began. A weather station in China’s Xinjiang recorded temperatures of 50 degrees Celsius in July.

The World Meteorological Organization’s State of the Climate in Asia report said the rate of temperature increase in Asia over the past 35 years was nearly twice that of the previous three decades.

The sight of potatoes effectively being ‘cooked’ underground in South Korea, therefore, is not merely an agricultural disaster in one country. It is also a stark warning of the rapidly intensifying climate risks facing Asia.

//DBTech/independent/MIS//

Speaking of the device, James Li, CEO of HONOR, opined, “Hardware sets the foundation for image quality, while algorithms deliver the leap. But technology is only the foundation, what ultimately defines great imaging is aesthetics.”

He added, “Together with ARRI, we will continue to push the boundaries of creation and inspire a new generation of creators.”

A camera that moves like a robot
The phone’s biggest attraction is its automated pop-up camera gimbal, designed to make photography and videography considerably easier. Made of titanium, the gimbal can rotate through 360 degrees and move in different directions, enabling shots that are difficult to capture with a conventional smartphone. It includes several cinematic shooting modes, such as Tilt Lock, FPV View and SpinShot.

The device also features an integrated AI algorithm for subject tracking. The camera can automatically follow a person’s movements by panning and tilting. Users can also open the camera arm and take photos using hand gestures.

Dual 200-megapixel camera system
The camera setup includes a powerful combination of a 200-megapixel main camera and a 200-megapixel telephoto lens. Through a partnership with professional cinema-camera manufacturer ARRI, the phone offers AI-assisted noise reduction, a wide dynamic range and RAW file output with real-time preview.

AI-powered ‘robot mode’
The phone runs on the MagicOS operating system and introduces a feature called ‘YOYO Robot Mode.' It combines multimodal perception, contextual understanding, gesture recognition and physical movement.

As a result, the gimbal can dance to music, nod in response to approval, or rotate according to voice commands.

Honor is also opening the platform to third-party developers, allowing them to use the phone’s multimodal capabilities and embodied interaction features to create new user experiences.

Flagship hardware
Beyond its robotic camera, the device comes with high-end smartphone specifications. It is powered by the Snapdragon Elite Gen 5 processor and features a dedicated cooling system and a large 7,060mAh battery.

The phone has a 6.3-inch OLED display with adaptive refresh-rate support and a peak brightness of up to 6,800 nits.

In China, pre-orders are currently available in two configurations: 12GB RAM with 512GB storage and 16GB RAM with 1TB storage.Official deliveries are scheduled to begin on August 1.

The 12GB version is priced at approximately $1,480, while the 16GB version costs around $2,075. 

While expressing excitement over the unveiling of the new phone, Honor Bangladesh did not immediately disclose when it would arrive in Bangladesh or how much it might cost.
//DBTech/engadget/BMT/OR/IH//

Attractive design and improved display

The Xiaomi Redmi Note 17 5G features noticeable design changes compared with its predecessor. Its flat frame and attractive rear finish give the handset a modern appearance. Gorilla Glass 7i has been used for display protection, while an IP65 rating provides protection against dust and light splashes of water.

The phone features a huge 7.0-inch AMOLED display with a resolution of 1080 × 2396 pixels and a 120Hz refresh rate. The screen is sufficiently bright, reaching a peak brightness of up to 1,800 nits. However, because of its large display and battery, the phone weighs 225 grams, which may feel somewhat heavy when used with one hand. An optical fingerprint sensor is also placed under the display for security.

Processor and software performance

For processing power, the phone uses Qualcomm’s Snapdragon 4 Gen 4 processor. Although it is somewhat less powerful than the processor used in the previous Redmi Note 15, it delivers smooth performance for everyday tasks, browsing, social media use and running regular apps.

The phone comes in 6GB and 8GB RAM variants, with 128GB and 256GB storage options. On the software side, it runs Android 16 with Xiaomi’s proprietary HyperOS 3 user interface. The new interface is reasonably fast and offers a clean and user-friendly experience.

Camera and sound system

For photography, the Redmi Note 17 5G has a single 50-megapixel main camera on the rear. Although it lacks optical image stabilization (OIS), the camera can capture fairly good and clear images in adequate lighting. In low-light conditions, however, some limitations become apparent in fine details and dynamic range. An 8-megapixel front camera is provided for selfies and video calls.

Due to limitations of the processor, video recording is restricted to a maximum of 1080p at 30 frames per second. Video footage can also show noticeable shaking when recording while walking. For audio, the phone has a single mono speaker at the bottom, which produces average-quality sound.

Massive battery and charging speed

The handset’s biggest selling point is its enormous 8,000mAh battery. Such a large battery is uncommon in conventional smartphones. Thanks to its relatively power-efficient processor, the phone can easily last two to three days on a single charge under moderate everyday use. This could be particularly useful for people who spend their days outdoors using ride-sharing services, delivery platforms or navigation apps.

The phone supports 45W fast charging and comes with a fast charger. Using the supplied charger, the battery can be fully charged in around one and a half hours. It also supports 22.5W reverse charging, allowing users to charge other devices from the phone.

Final verdict

Overall, the Xiaomi Redmi Note 17 5G appears to be designed for a specific group of users. Those who prioritize long battery life, a large display and reliable everyday performance may find it an excellent choice.

However, users looking for heavy gaming performance, professional-grade photography and videography, or stereo sound may notice the limitations of its processor and single-camera setup. Even so, considering practical needs and budget constraints, the phone remains a strong contender for users seeking a smartphone with exceptionally long battery life.

Three key takeaways

• ???? 8,000mAh battery: Can provide roughly two to three days of moderate everyday use.
• ???? 7-inch 120Hz AMOLED display: Excellent for watching videos and consuming other content, though the 225g weight makes one-handed use less comfortable.
• ⚙️ Strong practicality, limited versatility: Good for everyday tasks and long battery life, but less suitable for heavy gaming, advanced photography or high-quality audio.

//DBTech/BMT/OR//

The new phones are powered by Google’s latest Tensor G6 processor, which is around 20% more power-efficient than the previous Tensor G5. It also delivers 25% faster browsing and 15% faster app loading. As a result, the built-in Gemini Nano AI can perform complex tasks, including live translation, more quickly and efficiently. The camera module has also received a design overhaul, combining a metal frame with a smooth glass covering.

Major Camera and Display Upgrades

The Pixel 11 series brings significant improvements to camera technology. Faster image processing has increased low-light photography performance by around 4.5 times, a feature Google calls “Instant Night Sight.” The premium models also offer up to 120x Pro Zoom.

The standard Pixel 11 features a 6.3-inch OLED display with a peak brightness of 3,000 nits. The premium Pro models can reach up to 3,600 nits and are twice as scratch-resistant as last year’s models.

The standard Pixel 11 comes with a 48-megapixel main camera, a 13-megapixel ultrawide camera and a 10.8-megapixel telephoto camera.

The Pixel 11 Pro and Pro Fold introduce a completely new feature called “Highlight.” It is a circular LED light positioned beside the rear camera that illuminates when a call comes from a preferred contact or when Gemini AI is performing a task, even when the phone is placed face down.

The camera app also introduces “Magic Capture,” which can automatically take and enhance photos and videos with a single click. The “Camera Looks” feature allows users to apply professional artistic styles to their photographs.

Google has also introduced an accessibility feature for people with hearing impairments that can convert American Sign Language performed in front of the camera directly into text.

Longer Battery Life and Faster Charging

Battery and charging performance have also received major upgrades. Thanks to adaptive battery technology, the new phones can operate for more than 30 hours on a charge, while the foldable Pro Fold is rated for around 24 hours.

Wireless charging is up to 25% faster with the new PixelSnap magnetic charging system.

Higher Prices, More Storage

Because of increased component costs, the prices of all four models have risen by $100.

• Pixel 11: starting at $899
• Pixel 11 Pro: starting at $1,099
• Pixel 11 Pro XL: starting at $1,299
• Pixel 11 Pro Fold: starting at $1,899

However, the base storage of the standard and Pro models has been increased from 128GB to 256GB.

All four smartphones run on the new Android 17 operating system and offer IP68-rated water and dust resistance.

Pixel Watch 5 and Pixel Tag Also Unveiled

Alongside the smartphones, Google unveiled its new smartwatch, the Pixel Watch 5, and Pixel Tag, a tracking device designed to compete with Apple’s AirTag.

Powered by Qualcomm’s Snapdragon W5 Gen 2 chipset, the Pixel Watch 5 is 20% faster than its predecessor. It also introduces advanced health-monitoring capabilities, including long-term monitoring of blood-pressure patterns and indicators of insulin resistance.

The 41mm version offers up to 30 hours of battery life and starts at $399.

Meanwhile, the $30 Pixel Tag uses ultra-wideband and Bluetooth technology to help users locate their belongings. It is water-resistant, including underwater use, and can operate for an entire year on a CR2032 coin-cell battery.

//DBTech/BMT/OR//

To alleviate the suffering and physical travel hassles of bailed individuals and their families, the government has launched the online E-Bail Bond system in 12 additional districts. Law, Justice and Parliamentary Affairs Minister Md. Asaduzzaman inaugurated the expansion this morning on August 13, 2026 (Thursday) at a ceremony held in the ministry's conference room. With these 12 new districts, the digital legal service is now operational in 28 districts across Bangladesh.

The newly included districts are Tangail, Kishoreganj, Munshiganj, Cox's Bazar, Chandpur, Rangamati, Sirajganj, Bagerhat, Patuakhali, Habiganj, Thakurgaon, and Jamalpur.

Md. Asaduzzaman stated, “Under the conventional method, filing a bail bond required passing through 12 steps, whereas the E-Bail system accomplishes it in just two. Lawyers can now submit bail bonds directly via smartphones from anywhere, with instant OTP verification.”

The Law Minister highlighted that around 120,000 E-Bail bonds were submitted online in the initial 16 districts over recent months, demonstrating the system's effectiveness. This paperless, digital approach substantially reduces the need for repeated visits to prisons, magistrate courts, and district judge courts while enhancing transparency, accountability, and administrative speed.

Through a special power-management procedure called “Big Bang,” NASA has reorganised the spacecraft’s power system so that it can continue collecting scientific data through 2027 without shutting down any additional scientific instruments.

Fighting a Half-Century Power Decline

Launched in 1977, Voyager 2 is powered by special generators that convert heat produced by the radioactive decay of plutonium into electricity. After decades of operation, its plutonium supply has gradually declined, causing the spacecraft to lose about 4 watts of power-generating capacity every year.

Voyager 2 originally operated with 10 scientific instruments. To conserve power, NASA gradually switched off most of them, eventually reducing the number to three.

Earlier this year, the spacecraft faced the possibility of permanently shutting down another instrument because of its worsening power shortage. However, the Big Bang operation allowed NASA to simultaneously switch multiple power-consuming systems to lower-power alternatives while maintaining sufficient heat to keep the spacecraft functioning.

A Historic Journey Beyond the Solar System

Voyager 2 began its journey about two weeks before its twin spacecraft, Voyager 1, with the mission of studying the outer gas-giant planets of the Solar System. It reached Jupiter in 1979 and went on to achieve the unique feat of visiting all four of the Solar System’s giant planets within the following decade.

In 2018, Voyager 2 crossed the heliosphere—the protective bubble surrounding the Sun and its planets—and entered interstellar space. From there, the spacecraft continues to send valuable scientific data back to Earth about the density and temperature of plasma in deep space.

After nearly half a century, one of humanity’s most successful spacecraft is still answering questions about the space beyond our Solar System.

//DBTech/BMT/OR//

Initially released worldwide as a preview version, the new app gives Linux users access to ChatGPT, ChatGPT Work and Codex, OpenAI’s specialised technology for programming assistance.

OpenAI said the app will work directly on the desktop editions of Ubuntu 24.04 and 26.04 LTS, Debian 13, and Fedora 43 and 44. Although Linux has numerous distributions, the app can also be run on other compatible flavours and downstream distributions based on these platforms.

OpenAI Follows Anthropic

Although OpenAI has arrived somewhat later than rival AI company Anthropic, it has now brought the same convenience to its Linux users. About a month ago, Anthropic launched its Claude desktop app for Linux, supporting Ubuntu 22.04 and later and Debian 12 and later.

With the launch of its own Linux desktop app, OpenAI is expected to make it easier for Linux developers and technology enthusiasts to access ChatGPT and Codex under one platform, while further expanding access to the broader AI ecosystem.

//DBTech/BMT/OR//

Catching Up With ChatGPT and Search AI

Gemini’s rapid growth has intensified its direct competition with OpenAI’s ChatGPT in the artificial intelligence market. ChatGPT entered the 1-billion-user club last June. Google has been deeply integrating Gemini with Search, Google Workspace and Android.

Separately, Google’s AI Mode in Search is also being actively used by more than 1 billion users.

How Users Are Using Gemini

Google has highlighted several notable figures about user engagement with the Gemini app:

• Voice dominates: Around 63% of app users interact with Gemini directly through voice.
• Image generation: More than 150 million images are being generated through Gemini worldwide every day.
• iOS users: Gemini has more than 100 million active users on Apple’s iOS platform alone, in addition to its Android user base.

As AI technology continues to evolve, Google has also introduced the Gemini 3.5 Flash model to make coding and autonomous AI-agent tasks easier.

According to figures for the second quarter of 2026, Gemini had 950 million users before rapidly climbing to the 1-billion mark.

Google is also expected to announce more Gemini-powered features for Pixel devices at its upcoming Made by Google event.

//DBTech/BMT/OR//

‘Democratisation’ of Cybercrime and an Uneven Battle

In the past, launching a major cyberattack required considerable technical expertise and weeks of research. Now, the same tasks can potentially be accomplished within hours using an advanced large language model (LLM). Moreover, cybersecurity is inherently asymmetric. A security expert must identify and close every possible vulnerability or digital loophole, whereas a hacker needs to find just one hidden flaw or “zero-day vulnerability” to launch an attack. AI has made it considerably easier to identify such previously unknown software vulnerabilities.

Rise of Social Engineering and Phishing Attacks

One of the biggest targets for cybercriminals is human error and lack of awareness among ordinary users. In the past, phishing emails could often be identified through grammatical mistakes or inconsistencies. With AI, however, criminals can now create highly fluent, accurate and personalised “spear-phishing” messages targeting specific individuals.

Various studies have found that the use of AI in phishing emails has pushed the rate at which users click on fraudulent links sharply upward—from 12% to around 52%. At the same time, fraud involving voice cloning and deepfakes has also increased several times over.

AI Agents Create New Security Risks

To speed up digital tasks, organisations and individuals are increasingly using AI agents or automated assistants, such as OpenClaw. These agents are often given extensive administrative access to email, calendars, financial accounts and internal systems. However, because AI agents have fundamental security limitations, they are becoming easy targets for cybercriminals.

Through “prompt injection” attacks, malicious instructions can be hidden inside website code or messages. AI agents may unknowingly follow those instructions and end up leaking an organisation’s sensitive information.

Defenders Fight Back

Although cybercriminals are rapidly adopting new technologies, security professionals are not giving up. According to a Trend Micro survey, preventing deepfakes and application-level attacks such as prompt injection is being treated as a major priority.

Technology companies such as Microsoft are also developing agent-based AI security systems such as “Project Perception.” These systems involve red teams for penetration or hacking tests, blue teams for risk assessment and green teams for prevention, with the aim of automatically preparing for and blocking potential attacks.

Overall, the rapid rise of AI has turned the cyber world into a never-ending race. For the time being, the opportunities available to attackers appear to have gained a slight edge over those available to defenders.

//DBTech/BMT/OR//

Nvidia has signed memorandums of understanding with six leading Wall Street investment firms—Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR.

Under the new framework, the investment firms will create dedicated capital pools to finance purchases of Nvidia infrastructure. This will allow frontier AI labs, enterprises, government agencies and cloud service providers to finance the Nvidia chips and infrastructure they need at relatively lower interest rates and on more manageable terms.

Nvidia CEO Jensen Huang said the company itself has the capacity to provide a backstop for up to 25% of the initiative, equivalent to around $125 billion, to support its successful implementation.

According to Huang, chips and GPU technology have now moved beyond the traditional boundaries of semiconductors to become long-term, revenue-generating assets. Just like electricity and the internet, AI computing has become a fundamental form of infrastructure. He said the initiative would accelerate the development of a new generation of “AI factories” around the world to meet the demands of future industries.

Global AI budgets among major technology companies are expected to exceed $730 billion this year. However, given the enormous capital requirements of AI infrastructure, bringing in private-equity firms and institutional investors beyond government banks and capital markets could permanently reshape the global chip market. Although the agreements have been reached, the release of capital will begin gradually, subject to final approval of the deals.

//DBTech/BMT/OR//

Why Removable Batteries Disappeared

The growing consumer preference for slim phones, along with the use of premium metal and glass bodies, was a major reason removable batteries declined. Repeatedly opening a glass back cover can be risky. In addition, to protect devices from water and dust and achieve ratings such as IP68, most brands opted to make phones completely sealed.

Another factor was business-related: when batteries could not be replaced, consumers were more likely to purchase new phones sooner, which was beneficial for manufacturers. However, some rugged phones and environmentally friendly devices have continued to offer replaceable batteries.

New European Union Regulations

The situation is beginning to change because of a new European Union regulation. From February 2027, all smartphones and tablets sold in the European market will be required to have batteries that ordinary users can easily replace.

The main objectives of the regulation are to reduce electronic waste, encourage the recycling of valuable minerals and extend the lifespan of devices. Just as the adoption of USB-C has influenced smartphone design worldwide, manufacturers may bring back this design across their global smartphone lineups.

How Much Will the Design Change?

Under the regulation, “easily replaceable” means that users should be able to remove the battery using commonly available tools, such as an ordinary screwdriver. No heat or special chemicals should be required to loosen adhesive.

However, brands may receive special exemptions for flagship or premium phones if the batteries offer greater durability and the devices provide advanced water-resistance features, with battery replacement carried out by professional repair technicians.

As a result, the backs of all smartphones may not suddenly become removable in the old-fashioned way. However, over the coming years, replacing a smartphone battery is likely to become considerably easier than it is today.

//DBTech/BMT/OR//

Display-Free Tracker and Round Display

One of the most interesting ideas under consideration is a special fitness tracker without a display. Apple is reportedly exploring this concept to compete with fitness brands such as Whoop and Google’s Fitbit Air. The company has also been evaluating smartwatches in different shapes and display categories, including a design for a round Apple Watch. However, analysts believe the round version may ultimately never reach the market.

Premium and Affordable Versions

As part of expanding the lineup, Apple is considering launching a new, more affordable smartwatch instead of relying solely on the existing Apple Watch SE as its budget option. At the same time, the company may introduce a new model in the ultra-premium category. According to the report, Apple could bring back its popular “ceramic body” casing version by the end of this year or sometime in 2027.

New-Colored Bands and Features

Apple is also preparing new-colored straps or bands and different color options to diversify the user experience. With smart rings and screen-free fitness bands becoming increasingly popular, Apple appears to be significantly restructuring its smartwatch portfolio to maintain its traditional market and meet the needs of a wider range of customers.

//DBTech/BMT/OR//

Engineering & Technology

Ownership and Production Timeline
According to the report, Sony Group will hold about 60 percent of the planned joint venture, while TSMC will own the remaining 40 percent. The new factory will be built in Japan’s southern Kumamoto prefecture. If everything proceeds as planned, commercial production of sensor chips could begin at the facility by 2029.

Technology Partnership and AI Potential
The two companies had previously expressed interest in establishing a joint venture in Japan in May. At the time, they discussed combining Sony’s expertise in advanced sensor design with TSMC’s powerful manufacturing technology. Through the new agreement, the companies plan to create major opportunities not only in smartphone technology but also in automobiles powered by physical artificial intelligence, autonomous vehicles and robotics.

A New Market Equation
Sony has long dominated the global image sensor market for smartphones and automobiles. Market analysts believe the deeper partnership with TSMC will play an important role in helping Sony face competition from rivals such as Samsung and OmniVision. However, Sony declined to comment on the Nikkei report, while TSMC also did not immediately provide any comment.

//DBTech/BMT/OR//

According to sources familiar with the matter, Apple has begun preliminary discussions with CXMT with the aim of using its chips in selected devices sold exclusively in the Chinese market. CXMT, or ChangXin Memory Technologies, is currently China’s largest chip manufacturer by market value. However, neither international news agency Reuters nor Apple and CXMT immediately commented on the matter.

A New Source to Address the Shortage

The enormous demand for memory chips from AI-powered servers and data centers has tightened the global supply of processors and memory components for conventional electronic products. As a result, alongside the world’s leading chip suppliers, the US tech giant is turning to local manufacturers as an additional source of supply. This is not the first time CXMT chips have been used by major technology companies. Laptop manufacturers HP and Acer have already begun using CXMT memory chips in products sold outside the United States.

Regulatory and Technological Challenges

Existing trade tensions between China and the United States could make such a chip deal particularly complicated for Apple. Under US export-control restrictions, Apple cannot directly exchange certain technical information with CXMT. As a result, rather than having CXMT manufacture customized chips specifically for iPhones, Apple is reportedly exploring the use of CXMT’s existing standard memory chips.

At the same time, Apple is facing pressure from US lawmakers to avoid increasing its dependence on Chinese companies.

CXMT is already planning to build its second factory in Beijing to increase production. According to trade analysts, securing a major customer such as Apple could pose a significant challenge to the dominance of other Asian technology giants, including Samsung and SK Hynix, in the global memory-chip market.

//DBTech/BMT/OR//

OnePlus is offering users in the United Kingdom, Germany, Spain and France a free 50-watt AirVOOC magnetic wireless charger. However, customers must meet two specific conditions to receive the gift. First, they must visit OPPO’s German website and register for updates or subscribe to its newsletter. Second, they must link their existing OnePlus account to an OPPO account.

Limited Gifts and Additional Discounts

The offer is not available to everyone. A total of only 5,600 wireless chargers have been allocated across the four countries. There is no separate quota for any particular country, so the chargers will be distributed on a first-come, first-served basis. Interested customers have therefore been advised to complete the process through OPPO’s designated website as soon as possible.

OPPO is not limiting its farewell initiative to free wireless chargers. The company is also offering various discounts on products other than smartphones. The availability and amount of these discounts may vary from country to country.

The initiative is essentially aimed at ensuring that OnePlus customers do not face difficulties following the brand’s departure and can transition smoothly to OPPO’s services.

//DBTech/BMT/OR//

SSD vs. Pen Drive

In terms of speed and storage capacity, SSDs, or solid-state drives, are far ahead. A typical pen drive generally has a capacity of around 128GB, whereas SSDs can store one terabyte or even more. Because of their higher speeds, professional IT users and content creators have increasingly been turning to SSDs. In addition, their durability and the convenience of using USB Type-C ports on newer laptops have contributed to the growing popularity of SSDs.

AI Has Driven Up Prices

The concern that pen drives were becoming obsolete has been slowed largely by the sudden expansion of artificial intelligence. As demand for memory and storage in data centers has multiplied, SSD prices have risen dramatically worldwide. Available data show that the price of a good-quality 4TB SSD has increased by more than threefold in recent times. In contrast, pen drive prices have remained relatively low and stable. As a result, pen drives have once again become an affordable and practical option for ordinary users who only need to store documents and other essential files. According to Google Trends data, online searches for pen drives have also increased significantly in recent times.

Still Indispensable for Certain Uses

Pen drives have survived not only because of their low cost but also because of their usefulness in specific situations. There is still no real substitute for a pen drive when creating a bootable drive to install Windows or another operating system on a new computer. They are also useful for carrying computer repair tools and can easily be attached to a key ring for convenient portability. Therefore, even in the era of high-speed SSDs, pen drives are unlikely to disappear from the market anytime soon.

//DBTech/BMT/OR//

Not for Humans, but for AI

The browsers we normally use come with themes, tabs and various extensions. But an AI agent has little need for such visual features.

Cloudflare says its new browser has been designed primarily to focus on context-window management, improved performance and reduced token consumption. It is also capable of addressing security risks such as prompt-injection attacks. Using KiteSurf, software developers can now more easily build AI systems capable of automatically accessing websites, filling out forms and performing various browser-based tasks.

More Efficient and Cost-Effective Than Chromium

Perhaps the most striking detail is that Cloudflare developed the browser in just 12 weeks. It runs on the company’s serverless platform, Workers. For tasks such as taking screenshots and collecting information from websites, KiteSurf uses significantly less CPU and memory than the popular Chromium browser engine. Investigations by several international technology news outlets found that it consumes roughly three to seven times less memory and CPU than Chromium, potentially reducing developers’ server costs considerably. It is currently available free of charge in beta through the Browser Run service.

A Combination of Different Technologies

Several modern technologies were used to build the browser from the ground up. These include Blitz’s modular rendering engine, Firefox’s CSS parser Stylo and the Rust-based ECMAScript engine Boa JS. Cloudflare says that despite being new to the market, KiteSurf has already successfully passed tests involving more than 215,000 web platforms. Popular websites such as Wikipedia and Hacker News can be accessed easily through the browser. The company also plans to eventually make the browser available to everyone as an open-source project.

//DBTech/BMT/OR//

NextSlide founder Ahmed Bashir recently confirmed the deal in a message posted on the company’s website. He said their primary goal had been to make visual communication more accessible, allowing anyone to present their ideas clearly and effectively to others. Bashir said that after joining OpenAI, they would continue working toward the same goal.

According to him, they aim to develop AI products that help people create new things and turn their ideas into meaningful work.

Financial Details of the Deal Remain Unknown

Interestingly, OpenAI’s acquisition of the startup actually took place earlier this year. However, in a post on LinkedIn, a professional networking platform, Ahmed Bashir said the deal was being officially announced several months after the acquisition. Neither party has disclosed how much OpenAI paid to acquire NextSlide.

A History of Successful Ventures

This is not Ahmed Bashir’s first experience of building and selling a startup. He was previously a co-founder of another startup called Caper AI, which developed technology for smart shopping carts and automated checkout systems. In 2021, another company, Instacart, acquired Caper AI for a substantial amount. Now, Bashir’s second venture, NextSlide, has joined forces with a major company like OpenAI.

A New Horizon for ChatGPT

Technology experts believe the acquisition could enable ChatGPT to move beyond text and writing and become a powerful tool for creating presentations, competing with platforms such as Microsoft PowerPoint and Google Slides.

Currently, millions of professionals use chatbots to generate content and then move to separate software to create presentation slides. If slide-generation capabilities are integrated directly into ChatGPT, their workflow could become significantly faster. People familiar with the matter believe OpenAI is pursuing such initiatives not merely to develop text-based models, but also to gain greater control over users’ overall work experience.

//DBTech/BMT/OR//

According to a new report by market research firm Sensor Tower, India's mobile app market generated a record $345 million in revenue in the second quarter of this year, up 35 percent from the same period last year.

Key Drivers of Revenue

Sensor Tower said generative AI, streaming and productivity apps contributed more to this revenue growth than gaming. Indian consumers are increasingly willing to pay for digital subscriptions and premium services. Over the past three and a half years, revenue per app download in India has more than doubled. However, the number of app downloads per quarter has remained stuck at around 6.3 billion since 2023.

According to Sensor Tower analyst Eve Chen, the growing use of digital payment systems in India, particularly UPI and digital wallets, has made in-app purchases much easier.

India in the Global Context

India's growth is particularly notable compared with other countries. Among the world's leading app markets, India recorded the fastest revenue growth in the second quarter. While app revenue increased by 30 percent in Mexico and 25 percent in Turkey, revenue from apps in the United States declined by 3 percent.

India, however, still lags far behind mature markets such as the United States, South Korea and Japan. In the United States, each app download generates an average of $4.60 in revenue, while the figure remains very small in India. Nevertheless, the upward trend in revenue points to significant long-term potential.

Generative AI and Other Apps Take the Lead

Generative AI has emerged as the fastest-growing app category in India in terms of revenue. During the second quarter, apps such as ChatGPT and Claude accounted for about 83 percent of the total revenue generated by India's AI app sector.

According to data from another company, Appfigures, ChatGPT generates around $60,000 in revenue per day in India and was downloaded approximately 1.8 million times over the past month. Revenue from non-gaming apps has also increased significantly.

During the first half of this year, non-gaming apps accounted for 68 percent of India's total mobile app revenue. Consumer spending has also increased on platforms such as Google One, Amazon Prime Video, Sony LIV and JioHotstar. Meanwhile, although gaming revenue has declined globally, gaming revenue in India increased by 3.7 percent during the quarter.

//DBTech/BMT/OR//

In Bangladesh, the use of eSIMs has now become fairly common. Leading telecom operators such as Grameenphone, Robi and Banglalink have already introduced eSIM services for their customers. As a result, smartphone users are increasingly asking a common question: Should they choose an eSIM or a physical (traditional) SIM?

eSIM and Its Advantages

An eSIM is a digital SIM that is embedded electronically in a smartphone or wearable smart device, such as a smartwatch. There is no need to physically insert a SIM card into the phone. Mobile operators can activate a network connection or plan remotely on the device using wireless technology.

One of its biggest advantages is that it saves some space inside the phone. Smartphone manufacturers can use the space previously occupied by the SIM tray for other advanced components. eSIMs are also quite effective from a security perspective. If a phone is lost or stolen, someone cannot easily remove the eSIM to disconnect the device from the network.

For people who regularly travel at home and abroad, eSIMs can be particularly useful. Multiple eSIM profiles can be stored on a single phone, allowing users to switch between networks when necessary.

Physical SIM and Its Advantages

A physical or traditional SIM is the older technology that we have been using for years. Whether it is a mini, micro or nano SIM, the basic operating principle remains the same. Simply insert the SIM into the phone's SIM tray and it starts working.

Despite being an older technology, physical SIMs remain highly popular. Their biggest advantage is their simplicity. Suppose your phone suddenly stops working. You can easily remove the SIM from the damaged phone and insert it into another device. There is no need to visit the operator or go through the hassle of activating a new SIM. For people who frequently change phones, this convenience is difficult to give up.

Physical SIMs also have an advantage in terms of compatibility. Not all phones support eSIMs. From affordable feature phones to older smartphone models, standard nano SIMs can be used without difficulty.

Which One Should You Choose?

The number of eSIM-compatible smartphones in the Bangladeshi market is growing rapidly. If your phone supports eSIM, it can be a modern and secure option. It also offers an added advantage when it comes to protecting a stolen phone.

On the other hand, physical SIMs remain highly reliable. Particularly when a phone is damaged or a SIM needs to be transferred quickly to another device, there is still no substitute for a physical SIM.

Ultimately, there is no clear winner in this debate. The right choice depends on your usage pattern, phone model and personal convenience.

//DBTech/BMT/OR//

Starring Tom Holland and Zendaya, the blockbuster has become the 23rd film to surpass $500 million at the North American box office. At the top of that list is 2015’s ‘Star Wars: The Force Awakens’, which earned a total of $936.6 million. Previously, ‘Spider-Man: No Way Home’, released in 2021, took 12 days to reach the $500 million mark. By comparison, the new film’s seven-day achievement is being viewed as a major success in the film industry.

According to Hollywood trade publication Deadline, the film earned $32.7 million in North America last Wednesday alone. That makes it the second-highest-grossing non-opening Wednesday in North American box-office history. Once again, ‘Star Wars: The Force Awakens’ remains at the top with $38 million.

Film analysts expect the movie’s winning streak to continue through the weekend. Its revenue could rise even further with screenings in premium large formats and IMAX.

The film has already reached another major milestone worldwide, crossing $1 billion in global box-office revenue. Breaking one record after another, ‘Spider-Man: Brand New Day’ has now become one of the most talked-about and successful films of the year.

//DBTech/NMP/SME/OR//

Let's explore seven of the most discussed and effective platforms currently available, ready to turn your imagination into reality.

Midjourney
When it comes to AI image generation, Midjourney is often the first name that comes to mind. It's unparalleled in creating realistic and artistic images. It can produce perfect images, seamlessly blending light, shadow, and color exactly as you envision. Midjourney has shown remarkable skill, especially in generating portraits and fantastical scenes. It's used through the Discord app. By typing specific instructions or "prompts," this platform delivers four distinct image variations within seconds. However, it's not entirely free; its services require a subscription.

DALL-E 3
OpenAI has created an impressive image generation tool called DALL-E 3. Its direct integration with ChatGPT makes it incredibly easy to use. Users don't need to write complex technical instructions. Simply describe your ideas in plain language, and ChatGPT will organize them for DALL-E 3, generating the desired image instantly. Its biggest advantage is its ability to thoroughly understand every subtle detail of the instructions. From playfully drawn images to incredibly lifelike ones, it excels at all types of creative tasks.

Stable Diffusion
For those who desire more control over image creation, Stable Diffusion is an unparalleled choice. As an open-source technology, it's free to use and customizable. It can even be downloaded and used offline on powerful computers. Compared to other platforms, it offers significantly more image editing capabilities. However, a certain level of technical knowledge for writing prompts is required for effective use.

Adobe Firefly
Adobe, a familiar name in the design world, has introduced its AI, Firefly. Its biggest advantage is the assurance regarding copyright. Adobe trained this technology using its proprietary stock images, eliminating concerns about legal complications with generated images. Firefly seamlessly integrates with software like Photoshop and Illustrator, quickly earning the trust of professional designers.

Leonardo AI
Leonardo AI has become incredibly popular for creating images and characters in the gaming world. It's highly effective at generating 3D models or imaginative environments quickly and with great precision. Its user interface is very clean, making it easy for even new users to navigate. The ability to create a certain number of images daily for free also makes it attractive to general users.

Bing Image Creator
Bing Image Creator, developed by Microsoft, primarily functions using DALL-E technology. Its major advantage is that it's free to use with a Microsoft account. Accessing it is simple from any smartphone or computer through the Copilot feature. It has become a significant tool for everyday users to quickly generate high-quality images from simple text.

Canva Magic Media
Canva Magic Media is the proprietary artificial intelligence of the popular design website Canva. It's a magical tool for those who regularly create posters or images for social media. Users can directly write prompts within Canva to generate images and instantly integrate them into their main designs. This saves users a lot of time by eliminating the need to use separate platforms. Due to its user-friendly interface, it's regularly used by everyone from general users to small business owners.

With this incredible technological advancement, paintbrushes and colors are no longer needed to create images—only a good idea. And a few appropriate words to express that idea. These AI platforms are continuously improving. In the future, they might even surpass our imagination. But for now, anyone can become a digital artist using these tools.
//DBTECH/BMT/IH//

The program commenced with a one-minute silence in honor of the martyrs, followed by the reading of official messages issued by the President, the Prime Minister, the Foreign Minister, and the Minister of State for Foreign Affairs.

​During the discussion session, July warriors and youth representatives shared their firsthand experiences of the uprising. Speakers noted that the public's accumulated anger against prolonged fascist rule, corruption, misrule, deprivation of voting rights, and political oppression culminated in the mass student-people uprising in July.

They strongly demanded the extradition of the head of the fallen fascist Awami government to face trial for genocide and crimes against humanity, along with his associates. The discussion also highlighted the crucial role played by non-resident Bangladeshis in mobilizing international public opinion during the movement.

​In his concluding remarks, Consul General Kazi Muhammad Javed Iqbal paid tribute to the martyrs and emphasized that preserving the spirit and achievements of the historic victory against autocracy is a collective responsibility. He stressed that any conspiracy to destabilize the country and the government must be resisted at all costs through united efforts. He urged everyone to remain vigilant so that fallen fascist forces or any evil elements cannot disrupt the nation's peace, stability, and democratic progress. The event concluded with special prayers seeking eternal peace for the departed souls of the martyrs and wishing for the continued progress, peace, and prosperity of Bangladesh.

​//DBTech/MAM/SME//

Bots Are Changing the Internet
These puzzles, known as CAPTCHAs, have existed since the early days of the internet. However, with the rapid rise of cyberattack bots and advanced artificial intelligence, they have become significantly more difficult than before.

According to cybersecurity firm Imperva, bot traffic has increased dramatically over the past few years, even surpassing human activity on many parts of the internet. Many of these bots continuously scrape online content to train large language models, fundamentally changing the digital landscape.

Ordinary Users Caught in the CAPTCHA Trap
Websites are becoming increasingly desperate to defend themselves against automated attacks. Companies like Cloudflare conduct roughly 7.7 billion bot detection challenges every day. Yet bots have become so sophisticated that researchers at the University of Zurich in Switzerland found that machine learning systems can solve challenges such as Google reCAPTCHA with near-perfect accuracy.

As a result, websites have made CAPTCHAs even more complicated.

Instead of simply selecting traffic lights or bicycles, users may now be asked to choose two dice whose total equals seven or identify images showing main courses. These increasingly complex puzzles often frustrate ordinary users, particularly people with visual impairments.

One estimate suggests that people collectively waste the equivalent of 500 years every day solving CAPTCHAs worldwide.

Humans Are Losing the Battle
According to Professor Gene Tsudik of the University of California, humans have already lost the CAPTCHA arms race. To address the problem, some websites, including Reddit, now require users to log in before accessing older content.

However, if bots cannot be effectively controlled, websites risk being overwhelmed by automated traffic.

For that reason, these frustrating security checks are likely to remain. On today's internet, humans are becoming the minority, and proving that you're actually human may only become more difficult in the years ahead.
//DBTech/BMT/OR//

//DBTech/BMT/OR//

Not all gaming habits are unhealthy. Video games can help children develop problem-solving abilities, strategic thinking, and hand-eye coordination. These cognitive and motor skills are often viewed as positive outcomes when gaming is balanced with other aspects of daily life.

However, gaming becomes a concern when it begins to interfere with schoolwork, social interactions, sleep, eating habits, or other daily responsibilities. Parents should pay close attention to whether their children can regulate their own gaming time and how they react when asked to stop playing. Frequent anger, frustration, or an inability to disengage from games may indicate the early signs of addiction.

Experts suggest that strict punishment or suddenly confiscating digital devices is rarely an effective long-term solution. Instead, parents are encouraged to have open conversations with their children and establish clear daily routines that include reasonable screen-time limits. Allowing gaming only after schoolwork and other essential responsibilities have been completed can help children develop healthier digital habits.

Parents also play an important role by spending quality time with their children, reducing feelings of isolation, and encouraging alternative forms of entertainment. Outdoor sports, drawing, reading, cycling, and other creative or physical activities can naturally reduce children's dependence on screens. Family outings, shared activities, and participation in community events also contribute to healthy emotional and social development.

Rather than eliminating technology altogether, experts recommend teaching children how to use it responsibly and in moderation. With proper guidance and balanced routines, children can enjoy the benefits of digital entertainment while maintaining a healthy and well-rounded childhood.

//DBTech/BMT/OR//

Developed by Chinese tech firm Star.AI Spatio-Temporal Intelligence Technology Co., Ltd., the twin satellites will provide intelligent remote sensing services to Uzbekistan and Indonesia. Unlike conventional satellites that transmit raw data back to Earth for processing, these AI-driven units execute real-time on-board computing and data analysis in space. This slashes decision-making and disaster warning lead times from hours to just a few minutes.

Specifically, one satellite will monitor the entire lifecycle of Uzbekistan’s cotton production, optimizing crop management from sowing to harvesting. The second unit will support Indonesia's agricultural safety by assessing crop growth and mitigating natural disaster risks. China highlighted that integrating AI with space systems advances its high-tech capabilities, offering developing nations enhanced capabilities in agriculture, weather forecasting, mineral exploration, and disaster management.

Just hours before Russia launched its military offensive against Ukraine in 2022, a large-scale cyberattack disrupted a commercial satellite communications network across Europe. The objective was to cripple Ukraine's military and government communications infrastructure. The incident underscored growing concerns over the security of the global space economy, now valued at approximately US$626 billion.

Satellite Surveillance and Orbital Encounters

In recent years, incidents involving satellites from different countries approaching one another at unusually close distances in orbit have increased significantly. Analysts have likened these encounters to "dogfights" in space.

Russia's Luch-1 and Luch-2 satellites have reportedly maneuvered close to strategically important satellites such as Intelsat 39. Western intelligence agencies suspect these close approaches may be intended to gather sensitive information or test capabilities for potentially interfering with or taking control of other satellites.

Beyond cyber operations, several countries—including the United States, Russia, China, and India—have demonstrated anti-satellite (ASAT) capabilities by successfully destroying satellites with missiles.

However, experts warn that such tests generate thousands of pieces of orbital debris, which can remain in space for years and pose long-term threats to other operational satellites.

Nuclear Concerns and Defensive Strategies

Concerns have also intensified over the possible deployment of nuclear technology in space.

The United States has alleged that Russia is developing a space-based system capable of triggering a nuclear explosion in orbit. According to U.S. officials, such a detonation could generate a powerful electromagnetic pulse (EMP) capable of disabling large satellite constellations, including SpaceX's Starlink network.

Russia has denied the allegation, but the claims have further heightened international concerns about the militarization of space.

In response to these evolving threats, countries including France, Japan, Germany, and India are reportedly exploring the development of small "guardian" satellites designed to help protect high-value space assets.

Experts say outer space is no longer viewed solely as a domain for scientific exploration. It is increasingly becoming a new and strategically significant arena for geopolitical competition and future military conflict.

//DBTech/BMT/OR//

Choosing the right tool can help you complete complex tasks in minutes. Here are five powerful AI tools that are free to use and can significantly improve your productivity.

ChatGPT

ChatGPT is a versatile AI assistant that can help with a wide range of everyday tasks, including writing emails, drafting documents, explaining complex topics, brainstorming ideas, and creating project plans.

Whether you're overwhelmed with work or simply looking for quick answers, ChatGPT can provide information, organize your thoughts, and help streamline both personal and professional tasks. Its ability to generate well-structured content and solve problems quickly makes it an invaluable productivity companion.

Notion

Notion is an all-in-one workspace for organizing daily routines, managing tasks, taking notes, and planning projects.

It helps users eliminate clutter by keeping schedules, personal goals, study plans, and project information neatly organized in one place. With a clear overview of ongoing tasks, users can stay productive and complete their work more efficiently.

Grammarly

Clear and error-free writing is essential in professional communication, and Grammarly helps users achieve that effortlessly.

The AI-powered writing assistant automatically detects spelling, grammar, punctuation, and style issues while you write, offering real-time suggestions to improve your text. This allows users to produce polished, professional-quality emails, reports, and messages with minimal effort.

Otter

Taking detailed notes during meetings, online classes, or interviews can be time-consuming. Otter simplifies the process by automatically transcribing spoken conversations into written text.

The tool can also generate summaries of discussions, making it easier to review key points later. As a result, users can stay engaged during meetings without worrying about missing important information.

Gamma

Creating visually appealing presentations often requires considerable time and design skills. Gamma uses artificial intelligence to generate professional-looking presentations from simple prompts within minutes.

Whether for business, education, or personal projects, Gamma enables users to create polished slide decks quickly without spending hours on formatting and design.

The app, called Google AI Studio, was being developed for both Android and iOS. It had generated significant interest among developers and general users alike, with pre-orders available through both the Google Play Store and Apple's App Store.

Despite attracting nearly 800,000 registrations ahead of launch, Google has decided not to release the mobile app.

AI Studio Features to Be Merged into Gemini

According to Google's AI Studio team, the company wants to avoid asking users to install yet another standalone application. Instead, it has chosen to integrate AI Studio's capabilities directly into its flagship AI platform, Gemini.

Under this new strategy, both mobile and desktop Gemini users will gain access to the tools that were originally planned for the dedicated AI Studio app.

What the App Was Designed to Do

Google AI Studio was intended to let users build, test, and experiment with AI-powered applications using Google's Gemini models. It was designed as a development environment where developers and advanced users could create customized AI experiences without relying on separate infrastructure.

Technology analysts note that Google frequently develops innovative technologies across different internal teams before consolidating them into its flagship products. Rather than launching another standalone application, the company has decided to strengthen the already widely used Gemini ecosystem by incorporating AI Studio's functionality into it.

As a result, users who had been waiting for the mobile version of Google AI Studio will instead receive those capabilities through future updates to the Gemini app.

Google also confirmed that, while the mobile application has been cancelled, the web version of AI Studio will continue to operate as usual.

//DBTech/BMT/OR//

The newly integrated unit employs China's self-developed third-generation 'Hualong One' reactor technology. Over 90 percent of the unit's main equipment is domestically manufactured, delivering a power generation capacity of 1,200 megawatts. General Commander Zhang Guoqiang confirmed that operating both phase-one units at full capacity will yield over 18 billion kilowatt-hours of electricity annually, meeting residential and industrial demands for nearly 2 million people.

Under the master plan, the Taipingling facility will eventually house six 'Hualong One' reactor units. Once fully operational, the plant will produce around 55 billion kilowatt-hours of clean power annually, reducing standard coal usage by 16.52 million tons and cutting carbon dioxide emissions by roughly 58.2 million tons each year. Reporting credits belong to Xinhua, with editorial inputs from Nahar and Jennifer.

AWS and AI Fuel Record Growth

The sharp rise in Amazon's valuation followed the company's strongest cloud computing growth in four years and its announcement of increased annual capital expenditure (CapEx), signaling continued aggressive investment in AI infrastructure.

According to market analysts, Amazon is benefiting directly from its expanding partnerships with leading AI companies, including OpenAI, Anthropic, and Meta, particularly in cloud infrastructure and advanced AI chip deployment. These collaborations have strengthened AWS's position as one of the world's leading cloud platforms serving the rapidly growing generative AI ecosystem.

Founded by Jeff Bezos in 1994, Amazon reached a US$2 trillion valuation in June 2024. It has now added another US$1 trillion in market value in just over two years, underscoring investors' growing confidence in the company's long-term AI and cloud computing strategy.

Joining the Elite $3 Trillion Club

Amazon's latest achievement also highlights the widening gap among the so-called "Magnificent Seven" U.S. technology companies, as investors increasingly reward firms with strong AI-driven growth strategies and substantial investments in next-generation computing infrastructure.

With this milestone, Amazon becomes the fifth company to surpass the US$3 trillion valuation mark, joining Apple, Microsoft, Alphabet, and Nvidia in one of the most exclusive groups in global corporate history.

At present, Nvidia remains the world's most valuable publicly traded company, with a market capitalization approaching US$5 trillion.

//DBTech/BMT/OR//

Today, however, that simple joy is beginning to fade under the growing influence of artificial intelligence.

Where viewers once watched an animal video with delight and emotional warmth, many now instinctively ask a different question: Is this real—or was it generated by AI?

When Delight Becomes Artificial

According to Jessica Maddox, an Associate Professor of Digital Media at the University of Georgia, much of the emotional appeal of animal videos comes from believing that the events genuinely happened. Knowing that a puppy really behaved in a funny way or that a cat actually displayed an adorable reaction creates an authentic emotional connection between viewers and the living creature on the screen.

Artificial intelligence disrupts that connection.

Once people discover that a seemingly heartwarming video has been generated by AI, the emotional experience often feels hollow—or even deceptive. Instead of sharing a genuine moment with another living being, viewers realize they have been emotionally engaged by an algorithm.

The result is more than simple disappointment. Over time, it can make people increasingly skeptical of what they encounter online, gradually eroding trust in digital content.

Highly realistic AI-generated videos are already convincing enough to confuse large numbers of viewers. Many now scroll directly to the comment section—not to express their emotions, but to determine whether the footage is authentic. While some people may still enjoy these creations as harmless entertainment, experts warn that their broader implications could be far more serious.

A Warning of Greater Risks Ahead

Researchers caution that seemingly harmless AI-generated content can gradually normalize much more dangerous forms of digital deception.

If people become accustomed to watching fabricated videos of playful cats and dogs without questioning their authenticity, they may also become less skeptical of manipulated images of political leaders, staged crime footage, or other misleading visual content with far greater social consequences.

Another concern is the growing pressure on genuine content creators.

People who earn a living by producing authentic videos may find it increasingly difficult to compete with AI systems capable of generating virtually unlimited content around the clock. Unlike human creators, machines never tire, never need rest, and can produce countless videos within minutes.

As AI-generated media floods social platforms, building an audience—and earning a sustainable income through original content—may become significantly more challenging.

A recent example illustrates just how convincing these synthetic creations have become. A woman shared what appeared to be a charming video of several dogs skipping rope while wearing matching ribbons around their necks. The footage looked entirely authentic, but it was later revealed to have been generated by artificial intelligence.

Reflecting on the experience, she admitted that she never imagined she could be fooled so completely.

Incidents like this underscore a growing reality of the digital age: the internet is becoming a place where certainty is increasingly difficult to find. As AI-generated content grows more sophisticated, distinguishing truth from fabrication is becoming one of the defining challenges of the modern online world.

//DBTech/BMT/OR//

The shortage has been driven primarily by surging global demand for chips from artificial intelligence (AI) companies. While the crisis had previously affected lower-volume Apple products such as the Mac mini and Mac Studio, the company is now struggling to maintain normal supplies of its flagship MacBook Air. According to retailers, supply constraints for the laptop are more severe than ever before.

Longer Delivery Delays and Marketing Shift

Customers ordering a MacBook Air through Apple's official website are currently facing delivery times extending to late August, and in some cases into September. The company was also forced to postpone its annual student discount campaign, originally scheduled for June, due to the ongoing memory shortage.

Apple has also adjusted its marketing strategy. For the first time, the company is placing greater promotional emphasis on the more expensive 14-inch MacBook Pro instead of the MacBook Air. Advertisements now explicitly state that MacBook Air availability depends on existing inventory.

Alternative Supply Efforts and Market Outlook

To manage the situation, Apple has already increased prices in certain markets and begun seeking alternative memory chip suppliers worldwide. In particular, the company is exploring purchases from local memory manufacturers in China to help meet domestic demand.

Meanwhile, international semiconductor manufacturers are increasingly prioritizing the production of higher-margin memory chips for AI data centers over those used in conventional personal computers. As a result, the PC industry is facing growing supply constraints.

Technology analysts believe that with the rapid expansion of AI technologies, the global memory chip shortage is unlikely to ease in the near future.

//DBTech/BMT/OR//

Chinese technology companies have been aggressively advancing the development of open-weight AI models, aiming to deliver internationally competitive AI capabilities at lower costs. Alibaba's latest release further strengthens China's position in the global race for advanced artificial intelligence.

Among the World's Leading AI Models

Qwen 3.8-Max is built with 2.4 trillion parameters, making it one of the largest AI models ever developed.

According to the latest rankings published by the AI benchmarking platform Arena.ai, the model has become the highest-ranked Chinese-developed AI system for text analysis.

Globally, however, it trails only Anthropic's Claude Fable 5 and Claude Opus models. In image and visual content analysis, Qwen 3.8-Max ranks second worldwide, surpassed only by one version of Claude Fable 5.

Advanced Capabilities and Efficient Architecture

The new model is capable of simultaneously processing text, images, and video while supporting context windows of up to one million tokens in a single prompt.

This enables it to analyze massive software codebases, lengthy legal documents, or hundreds of pages of technical material in one session before generating comprehensive solutions.

To maximize both speed and cost efficiency, Alibaba has adopted a Mixture-of-Experts (MoE) architecture. Rather than activating the entire model for every request, only the specialized components relevant to a given task are engaged. As a result, although the model contains 2.4 trillion parameters, only about 95 billion parameters are activated for each inference, significantly reducing computational requirements.

Alibaba also stated that Qwen 3.8-Max successfully completed a complex software engineering project in just 16 days, demonstrating its advanced coding capabilities.

The company announced that the model will become available to users on the Alibaba Cloud Model Studio platform next week.

//DBTech/BMT/OR//

The findings were revealed in an investigative report published by the international news agency Reuters.

According to information compiled by the Washington-based Jamestown Foundation and an analysis of more than 80 Chinese academic papers and patents, institutions affiliated with the People's Liberation Army (PLA) have been using advanced U.S. AI models while circumventing existing American restrictions.

Model Distillation as a Technological Shortcut

The report states that Chinese military researchers are primarily relying on a technique known as model distillation. This process uses the outputs of a large and highly capable AI system to train a smaller, specialized model that can operate locally with far less computing power.

With the United States maintaining strict export controls on advanced semiconductor chips and related technologies, Chinese defense organizations reportedly view American AI models as both a source of technical insight and a shortcut to narrowing the technological gap with the United States.

The reviewed research papers suggest that Chinese researchers have gone beyond simply obtaining accurate answers. They have also sought to absorb the underlying reasoning processes and latent strategic logic employed by Western AI models.

Sensitive Code and Military Applications

The analysis indicates that PLA Unit 96941, a Beijing-based military intelligence and cyber warfare unit, has used OpenAI's models to process sensitive military source code.

Because third-party AI models are not considered suitable for directly handling classified information, researchers reportedly used the U.S. models to generate summaries of software code before transferring the distilled knowledge to a domestically developed AI model operating within a secure military network.

In addition, researchers at North China University reportedly used Anthropic's models to generate synthetic training data for social media monitoring and content moderation systems.

Meanwhile, the Academy of Military Sciences is said to have employed the technology to improve target identification capabilities for drones, naval vessels, and unmanned submarines during maritime exercises.

Rising Geopolitical Tensions

The revelations have intensified geopolitical tensions ahead of upcoming discussions between the United States and China on AI governance and security.

U.S. officials have alleged that such unauthorized use violates intellectual property rights and circumvents American export controls. Beijing, however, has rejected those accusations, countering that Washington is attempting to establish unilateral dominance in the field of artificial intelligence.

//DBTech/BMT/OR//

Neither his family nor official medical authorities have publicly released the exact cause of death, leading to heartfelt condolences alongside discussions across social media platforms.

Academic Brilliance & Research Legacy

Son of prominent BUET Civil Engineering Professor and earthquake specialist Dr. Tahmid Al-Hussaini, Irfan graduated from BUET’s Department of Electrical and Electronic Engineering (EEE). He then pursued higher studies in the US, earning both his M.S. and Ph.D. degrees in Electrical and Computer Engineering from Georgia Tech.

During his academic tenure, Dr. Irfan specialized in interpretable machine learning for healthcare, biomedical data analysis, document filtering, and integrating domain knowledge into AI models. His research papers were published in tier-1 international conferences including NeurIPS, SIGIR, CIKM, ICASSP, BioNLP, and MLHC.

Following his Ph.D., he joined NVIDIA as a Senior Data Scientist in Atlanta, focusing on healthcare AI applications.

An Inspiring Journey

In May 2024, Georgia Tech published an official story spotlighting Dr. Irfan and his wife, Dr. Fabia Farlin Athena. Although both attended BUET around the same time, they met at Georgia Tech, got married in Atlanta in 2022, and celebrated their Ph.D. graduation together in Spring 2024.

Remembered by BUET peers not only as a top-tier scholar and IEEE Signal Processing Cup contributor, but also as a skilled winger on the university football team and a deeply humble individual, Dr. Irfan’s untimely departure marks a tragic loss for Bangladesh's tech and scientific representation on the global stage.

//DBTech/FBP/SME//

Record Sales Performance

According to Reuters' calculations based on BYD's official sales statement released on Saturday, the company sold 419,211 vehicles in July, representing a 21.8 percent increase compared with the same month last year.

The July performance is being viewed as another major milestone for the company, reinforcing BYD's position as one of the world's fastest-growing electric vehicle manufacturers.

Exports Reach New High

The strongest growth came from international markets. BYD's overseas shipments—including passenger vehicles and pickup trucks—rose 124.3 percent year-on-year to a record 179,841 units in July.

The record-breaking export performance highlights the rapidly expanding global demand for BYD vehicles and underscores the company's growing international footprint.

Despite facing some challenges in China's domestic market, BYD has successfully offset weaker local demand through continued overseas expansion. Its sustained growth in the global electric vehicle market is further strengthening the company's position as a leading international EV manufacturer.

//DBTech/BMT/OR//

However, subsequent investigations by several international media outlets suggest the breach was even more severe. Within just 41 minutes, attackers reportedly drained approximately 70.2 million dollars' worth of Bitcoin from 1,196 separate wallets. The attack has been linked to a major security vulnerability in Coldcard, the Bitcoin hardware wallet developed by Canadian company CoinKite. All affected wallets reportedly contained at least 0.15 Bitcoin and were configured as single-signature wallets.

How the Attack Happened

Every cryptocurrency wallet relies on a seed phrase—a secret sequence of words that serves as the master key for recovering access to digital assets. These seed phrases are intended to be generated using highly secure random number generators, making them effectively impossible to predict.

However, a report by Block's Bitcoin Engineering and Security team revealed that Coldcard's firmware bypassed its dedicated hardware-based random number generator and instead relied on a weaker software algorithm.

According to the report, the wallet generated its secret values using Yasmarang, a weak pseudorandom number generator included in MicroPython. Because the generated values were predictable, attackers were reportedly able to reconstruct wallet seed phrases offline and ultimately gain access to users' Bitcoin holdings.

Who Is at Risk?

Coldcard is designed as an offline hardware wallet, allowing users to store Bitcoin without maintaining a direct internet connection. Owners of several older models—including the MK2, MK3, MK4, Q, and MK5—have reportedly been affected by the vulnerability.

CoinKite has stated that its newer hardware models are not affected by the flaw. However, the company warned that simply updating the firmware on older devices does not eliminate the risk associated with seed phrases generated before the fix. Users are therefore advised to update their firmware and create entirely new wallets with newly generated seed phrases.

The incident underscores growing cybersecurity concerns within the cryptocurrency sector. During the first half of this year alone, approximately 1 billion dollars (around 9,547 crore taka) worth of cryptocurrency assets were reportedly stolen, highlighting persistent weaknesses in digital asset security.

//DBTech/BMT/OR//

The company said it had made the difficult decision after careful consideration, confirming that the game's servers will be permanently shut down on 30 October.

NFT Hype Fades as the Game Ends

Although Champions Tactics maintained a dedicated community of players, many had anticipated the shutdown for some time. In May, Ubisoft announced that it would remove the game's Web3 functionality, a move widely seen as signalling the beginning of the end.

The game's primary attraction was its battle units, which could be bought and sold as NFTs on blockchain marketplaces using digital currencies.

Ironically, Champions Tactics was released in 2024, well after the peak of the NFT boom. NFT popularity had reached its height in 2021 before declining sharply throughout 2022. Ubisoft's decision to launch an NFT-focused game nearly three years after high-profile NFT promotions involving celebrities such as Paris Hilton and Jimmy Fallon had already faded surprised many industry observers.

Focus Shifts to a New Blockchain Game

Despite ending Champions Tactics, Ubisoft has not abandoned its blockchain gaming ambitions. The company is now attempting to migrate existing players to another title, Might & Magic Fates, by offering various in-game rewards.

The new game is a digital card battler similar to Hearthstone, where players will be able to buy, sell, and trade digital cards using either traditional currency or cryptocurrency.

//DBTech/BMT/OR//

Eye-Catching Design and the New Glowbar

The leaked images reveal three new Googlebook devices, all featuring the distinctive Googlebook logo on their keyboards. Two of them are laptops with a design closely resembling Lenovo's Chromebook Plus 14, while the third is a detachable 2-in-1 tablet equipped with a kickstand and a removable keyboard, reminiscent of the IdeaPad Pro Gen 2. Although the number of USB-C and USB-A ports varies between the models, all of the devices are finished in an elegant white colour.

The most striking design feature is a new illuminated strip called the Glowbar. Google is reportedly positioning the Glowbar as a signature visual element that will distinguish Googlebook devices from competing laptops. Similar lighting effects are also expected to appear in Google's upcoming Pixel smartphones.

Expanding Google's AI Vision

Previously, Google showcased only an unidentified laptop while promoting its new AI-powered platform. The leaked images now suggest that Googlebook branding will extend beyond traditional laptops to include tablets as well. While ChromeOS and Android have long supported tablets, the new platform appears to be designed for a broader range of form factors.

Most of the Googlebook features demonstrated so far rely heavily on mouse and voice interactions. One example is the new Magic Pointer feature, which enables users to control Google's Gemini AI simply by moving the cursor and issuing commands.

Google has yet to announce an official launch date for the Googlebook lineup. However, industry expectations point to a release later this year. Since the initiative is built around third-party hardware manufacturers, Googlebooks are expected to launch separately from Google's Pixel smartphone event. In addition to Lenovo, Asus, Dell, and HP are also expected to serve as hardware partners for Google's new platform.

//DBTech/BMT/OR//

How Personal AI Agents Will Be Used

According to Zuckerberg, future AI agents will simplify many aspects of everyday life by handling complex personal tasks. He said they will assist users with personal finance management, healthcare coordination, communications, and household organization, allowing people to delegate routine responsibilities to intelligent digital assistants.

He also believes that messaging platforms such as WhatsApp and Messenger will become the primary interfaces through which people interact with multiple AI agents. Meta revealed that more than one million businesses worldwide have already adopted its Business AI Agents on WhatsApp and Messenger.

Massive Investment in AI Infrastructure

To support the next generation of AI-powered services, Meta is making significant investments in infrastructure and computing capacity. As part of this strategy, the company has signed a major agreement with global investment firm BlackRock to help develop a $14 billion mega data center in El Paso, Texas, aimed at providing the enormous computing power required for advanced AI systems.

While competitors such as Google and Anthropic are focusing on AI-powered search and coding assistants, Meta intends to build its future business around personal AI agents, in addition to offering AI processing capacity as a commercial service.

Investor Concerns and Financial Impact

Despite Zuckerberg's ambitious vision, investors have expressed concerns over the scale of Meta's AI spending. Heavy investment in AI infrastructure and research caused the company's free cash flow to fall 91% year over year, dropping from $8.55 billion to just $784 million during the latest quarter.

Meanwhile, Reality Labs, Meta's virtual reality and VR headset division, reported a quarterly loss of approximately $4.6 billion, bringing its cumulative losses since 2021 to around $88 billion.

Following the release of Meta's financial results, the company's share price fell by roughly 10%. Nevertheless, Zuckerberg remains confident that personal AI agents will become Meta's next major product category and a key driver of the company's future revenue growth.

//DBT/BMT/OR//

Performance Benefits and Cache Management

Both Android and iOS devices are designed to remain operational for weeks without requiring a restart. To speed up app launches and improve responsiveness, smartphones automatically store temporary files in cache memory.

Over time, however, memory management can become less efficient, occasionally leading to issues such as app crashes, increased battery consumption, or sluggish performance. Restarting the device clears unnecessary temporary data from RAM, allowing the operating system to start fresh and often improving overall responsiveness. These performance improvements tend to be more noticeable on Android devices than on iPhones.

Cybersecurity and the BFU State

Many users believe that simply restarting a smartphone provides complete protection against cyberattacks. In reality, rebooting can temporarily disrupt certain advanced attacks, such as spear-phishing campaigns or zero-click malware, because malicious code stored only in memory is removed when the device powers down. However, determined attackers may attempt to compromise the device again after it restarts.

The more significant security benefit lies in what is known as the Before First Unlock (BFU) state. After a phone is restarted, all user data remains fully encrypted until the owner unlocks the device for the first time using a passcode or PIN.

If a phone is lost or stolen while still in the BFU state, accessing the data stored on the device becomes significantly more difficult, providing an additional layer of protection against unauthorized access.

How Often Should You Restart Your Phone?

For users who prefer a regular maintenance routine, restarting a smartphone once a week is generally sufficient.

Some smartphone manufacturers, including Samsung, offer built-in features that can automatically restart the device at scheduled times. Meanwhile, certain Google Pixel devices and iPhones automatically reboot after remaining unused for approximately 72 hours as an added security measure.

If your phone feels unusually slow or experiences minor issues such as temporary network connectivity problems, manually restarting it is often the quickest and simplest solution. For everyday use, however, there is no need to restart a modern smartphone every day.

//DBTech/BMT/OR//

Under the "ChatGPT for Academic Researchers" program, researchers from selected universities and research institutions will receive complimentary access to OpenAI's cutting-edge AI technology, along with dedicated technical support.

What Researchers Will Receive

Participants in the program will benefit from direct hands-on technical assistance from OpenAI and gain access to the company's latest flagship AI model, GPT-5.6 Sol Pro.

Each registered institutional account will also be able to add up to four additional collaborators from the same organization, enabling research teams to work together more effectively.

The program will launch this summer with 10,000 researchers, with OpenAI planning to expand participation to 100,000 researchers by 2027. The initiative builds on the success of ChatGPT Edu, the company's education-focused offering for academic institutions.

Privacy and Long-Term Goals

Addressing concerns over research confidentiality, OpenAI stated that, by default, researchers' confidential and private data will not be used to train its AI models.

The company aims to provide US$250 million worth of AI services to support external scientific research by 2027.

However, market analysts suggest that by integrating OpenAI's tools into researchers' daily workflows, the company could eventually strengthen its position in the commercial research sector and generate significant long-term revenue.

Expanding Its Presence in Academia

This is not OpenAI's first effort to engage the academic community.

Earlier this year, the company introduced Prism, an AI-powered tool designed to assist researchers with tasks such as verifying citations in scientific papers, formatting documents in LaTeX, and drafting academic manuscripts.

Industry observers believe the new free-access program could significantly broaden the adoption of OpenAI's AI technologies across the global scientific research community.

//DBTech/BMT/OR//

The report attributes the slowdown to rising memory chip prices and persistent supply constraints, which have prompted smartphone manufacturers to scale back the production of new handsets.

Qualcomm and MediaTek Take the Hardest Hit

The two leading chipset manufacturers, Qualcomm and MediaTek, suffered the steepest declines during the period.

Compared with the first half of 2025, shipments from both companies fell by approximately 25%.

Demand for Qualcomm's flagship Snapdragon 8 Elite Gen 5 and other premium processors remained relatively subdued. Meanwhile, MediaTek's affordable 5G chipsets recorded a sharp drop in sales, although its flagship Dimensity 9500 series maintained comparatively stronger market performance.

Apple, UNISOC, and Samsung Defy the Downturn

Despite the broader market slowdown, Apple, UNISOC, and Samsung recorded growth in both demand and chipset shipments.

• Apple: Strong demand for the new iPhone 17 lineup significantly boosted shipments of its A19 and A19 Pro processors.
• UNISOC: As smartphone brands increasingly turned to 4G processors for entry-level devices due to the higher cost of 5G chipsets, UNISOC capitalized on the trend. The company also expanded sales through new partnerships with Xiaomi's Redmi and POCO smartphone series.

Market Outlook

According to Counterpoint Research analysts, the overall decline in smartphone chipset shipments for the full year 2026 is expected to reach approximately 14%.

The impact is likely to be particularly severe in the entry-level smartphone segment, where chipset shipments could fall by as much as 30%.

Industry experts also warn that the memory chip market is unlikely to return to normal before the second half of 2027, meaning supply constraints could continue to affect smartphone production and pricing over the coming months.

//DBTech/BMT/OR//

Explore the Ocean Floor

The headline feature of the update is a vast new underwater area that resembles the ruins of the fictional Cerulean City. To access the new environment, players will gain a special Dive ability, allowing them to descend beneath the sea and explore the ocean floor.

The Dive feature will be released as a free update for all players, enabling exploration of underwater environments even for those who do not purchase the expansion.

The game's core objective remains unchanged: attracting different Pokémon by creating habitats suited to their individual preferences. However, the underwater setting introduces a new building mechanic that allows players to use floating blocks, making it possible to construct elaborate three-dimensional structures beneath the sea.

New Decorations and Expansion Pass

The Bubbly Basin expansion also adds a variety of underwater-themed furniture and decorative items, a watermelon cultivation feature, and a new Dream Island filled with useful resources.

One limitation, however, is that the new area cannot be purchased separately. Players must buy the Pokémon Pokopia Expansion Pass, priced at US$35, to access the DLC.

The Expansion Pass will also grant access to three additional major DLC releases planned for the future.

//DBTech/BMT/OR//

As part of the restructuring, AlphaFold's remaining researchers have been reassigned to other divisions within DeepMind, while several members of the original team have already left the company.

A Landmark Achievement in Science

Google DeepMind launched the AlphaFold project in 2018 to address one of biology's most enduring challenges—predicting the three-dimensional structures of proteins.

For nearly half a century, scientists struggled to determine how amino acid chains fold into complex protein structures. Traditional methods, such as X-ray crystallography and nuclear magnetic resonance (NMR), often required years to identify the structure of a single protein.

Using artificial intelligence, AlphaFold transformed the field by predicting protein structures with remarkable accuracy in just minutes.

The project later made structural predictions for the entire human proteome and nearly 200 million proteins freely available to researchers worldwide. The technology significantly accelerated research into drug discovery, vaccine development, and the study of neurological diseases such as Alzheimer's and Parkinson's.

In recognition of this landmark achievement, DeepMind Chief Executive Officer Demis Hassabis and senior researcher John Jumper were awarded the 2024 Nobel Prize in Chemistry.

Why the Team Was Disbanded

Despite the project's historic success, Google has decided to wind down the original AlphaFold research team less than two years after the Nobel Prize announcement.

Last month, AlphaFold's lead scientist, John Jumper, left Google DeepMind to join rival AI company Anthropic. Several experienced researchers reportedly departed alongside him.

The remaining members of the AlphaFold team have been reassigned to Gemini-related projects, enzyme design research, or Isomorphic Labs, DeepMind's drug discovery subsidiary.

Explaining the strategic shift, Pushmeet Kohli, Vice President of Research at Google DeepMind, said the company's focus over the past nine years had been to solve specific scientific problems. However, changing technological priorities have prompted a broader strategy.

Rather than concentrating on individual scientific challenges, Google now aims to dedicate its resources to developing increasingly capable, general-purpose Gemini AI models that can help automate the entire scientific research process.

//DBTech/BMT/OR//

Screen Sharing and Call Management from the Browser

The new Calls tab on WhatsApp Web allows users to view their previous call history and favorite contacts. Similar to the desktop and mobile applications, the web version now supports screen sharing, in-call reactions, and access to call history.

All calls made through the browser will remain protected by WhatsApp's end-to-end encryption, ensuring that conversations remain private and secure.

Seamless Switching Between Devices

In addition to web calling, Meta has introduced several improvements to the overall WhatsApp calling experience. One of the most notable features is the ability to transfer an ongoing call from one device to another without disconnecting.

For example, users who begin a call on their smartphone can seamlessly move it to a laptop or web browser upon arriving at home or the office, without ending the conversation.

Waiting Room and Enhanced Audio Quality

To improve security during online meetings, WhatsApp has introduced a Waiting Room feature. When creating a group call link, hosts can require participants to obtain approval before joining the call. This means that clicking the invitation link alone will not grant immediate access; participants will only be admitted after receiving the host's permission.

The platform has also introduced noise suppression technology to reduce unwanted background sounds, making conversations clearer in noisy environments. In addition, WhatsApp now delivers HD-quality video much earlier during video calls, improving the overall visual experience from the beginning of a conversation.

Meta also announced that the ability to place calls using a user's username, instead of a phone number, will be rolled out to all users in the near future.

//DBTech/BMT/OR//

Product Demand and Strategic Vision

At the opening of trading, Apple's stock rose by as much as 1.8 percent, reaching $342.89 per share. The gain allowed the iPhone maker to reclaim its position as the world's most valuable company earlier this month, overtaking Nvidia.

Apple's shares have climbed approximately 25 percent this year, outperforming many of Silicon Valley's leading technology companies.

While many of its competitors have spent billions of dollars building artificial intelligence (AI) infrastructure and data centers—putting pressure on their cash reserves and increasing debt—Apple has pursued a more measured strategy. Instead of investing heavily in developing its own large-scale AI models, the company has integrated Google's AI technology into its Siri platform, allowing it to avoid massive infrastructure costs.

A New Strategy to Retain Customers

According to market analysts, Apple's decision to keep iPhone prices unchanged, despite modest price increases for MacBook and iPad models, has helped strengthen consumer demand. Concerns over possible future price hikes have also encouraged more customers to purchase iPhones sooner rather than later.

On the same day, Apple launched a new iPhone leasing program in partnership with financial services provider Klarna. Under the initiative, customers can use iPhones through affordable monthly installment payments instead of paying the full purchase price upfront.

Market research firm Forrester believes the program is a timely move that eases consumers' psychological concerns about affordability without directly reducing iPhone prices, while also helping accelerate sales.

Overall, market analysts are optimistic that Apple will report revenue growth of more than 15 percent in its next quarterly earnings report.

//DBTech/BMT/OR//

'Four Plus Four' Flagship Strategy

According to the report, Samsung intends to divide its annual product launch schedule into two phases under a new "Four Plus Four" or "4+4" strategy. The company is expected to introduce four conventional flagship smartphones during the first half of the year, followed by four foldable flagship models in the second half.

Technology analysts believe the unprecedented move is designed to intensify Samsung's competition in the global premium smartphone market.

Galaxy S27 Series

Samsung's highly anticipated Galaxy S27 series is expected to debut in the first quarter of 2027. While previous Galaxy S series lineups typically featured three models, the upcoming generation is rumored to add a fourth variant.

Alongside the relatively affordable Galaxy S27 and Galaxy S27 Plus, and the top-tier Galaxy S27 Ultra, Samsung is expected to introduce the Galaxy S27 Pro for the first time.

Rumors suggest that the Galaxy S27 Pro will feature a 6.47-inch display and serve as a more compact alternative to the Galaxy S27 Ultra while offering premium features and a powerful camera system.

Expansion of the Foldable Lineup

In the second half of the year, Samsung is expected to unveil a new generation of foldable smartphones. The lineup is rumored to include the Galaxy G Flip 9, Galaxy G Fold 9, Galaxy G Fold 9 Ultra, and the second-generation tri-fold smartphone, the Galaxy G TriFold 2.

Recent rumors suggesting that Samsung might discontinue the Galaxy Flip series appear to have been disproved by reports pointing to the arrival of the Galaxy G Flip 9.

At the same time, Samsung has reportedly already begun preliminary development of the Galaxy G TriFold 2 and has initiated discussions with key component suppliers. The unique tri-fold smartphone is expected to retain its 9.96-inch main display while adopting a noticeably slimmer design than its predecessor.

//DBTech/BMT/OR//

Apple Silicon and Passive Cooling

The key reason Apple was able to eliminate the cooling fan is the exceptional power efficiency of its custom M-series chips. Compared with the Intel processors previously used in MacBook Air models, Apple Silicon generates significantly less heat while delivering higher performance with lower power consumption.

Instead of using an active fan, the laptop relies on a passive cooling system featuring an aluminum heat spreader. This mechanism transfers heat away from the processor and distributes it throughout the aluminum chassis, allowing it to dissipate naturally. As a result, the MacBook Air remains completely silent during everyday tasks.

Pros and Cons of a Fanless Design

The biggest advantage of a fanless laptop is silent operation. Unlike many gaming or high-performance laptops that produce noticeable fan noise under load, the MacBook Air operates without any mechanical sound.

Removing the fan and related cooling components also makes the laptop lighter and contributes to longer battery life.

However, the design has its limitations. During extended workloads such as 4K video editing, software compilation, or demanding gaming, the laptop can become noticeably warm. To prevent overheating, the processor automatically reduces its performance—a process known as thermal throttling. As a result, sustained heavy workloads may run slower than on laptops with active cooling.

Who Should Choose the MacBook Air?

The MacBook Air is an excellent choice for students, writers, office professionals, and everyday users who value portability, long battery life, and silent operation.

For professionals who regularly perform intensive tasks such as graphics design, 3D rendering, or high-resolution video editing, the MacBook Pro lineup—with its active cooling system and higher sustained performance—may be the better option.

//DBTech/BMT/OR//

Services Affected

The unexpected disruption prevented users from streaming music, searching for songs or albums, signing in to their accounts, and accessing playlists or account settings. Many users encountered error messages when attempting to use the Spotify app or website.

Thousands of users reported the outage on Downdetector, a website that monitors internet service disruptions, while many others took to social media to express frustration over interrupted music playback and inaccessible services.

Spotify's Response

After initially announcing that the first outage had been resolved, Spotify's official support account on X (formerly Twitter) reported another round of technical issues at around 3:00 p.m. The company said its engineers were investigating the problem and asked users for their patience while providing links to customer support resources.

Following approximately three hours of troubleshooting, Spotify confirmed that the technical issues had been resolved and that all services had returned to normal. Reports of outages on Downdetector also declined rapidly as users around the world regained access to the platform.

//DBTech/BMT/OR//

Direct Voice, Messaging, and Emergency Services

According to Amazon LEO, deployment of the second-generation satellite constellation is expected to begin in 2028. The primary goal is to provide voice calls, text messaging, mobile data, and emergency communication services directly to standard smartphones and other mobile devices in remote and underserved areas beyond the reach of conventional cellular towers.

Amazon currently has more than 390 broadband satellites operating in orbit and plans to launch its fixed broadband service later this year. The company has already secured partnerships with several international telecommunications providers, including Vodafone, DirecTV, HeroTel, and Australia's National Broadband Network (NBN).

Competition With SpaceX Intensifies

The satellite expansion places Amazon in direct competition with SpaceX in the rapidly growing space-based mobile connectivity market.

Earlier this year, SpaceX submitted an application to the FCC seeking authorization for a satellite network of up to one million satellites. Amazon urged the FCC to reject the proposal, arguing that the plan was unrealistic and based on speculative ambitions rather than a practical deployment strategy.

FCC Chairman Brendan Carr responded by criticizing Amazon's objection, saying on social media that the company should focus on advancing its own satellite program instead of attempting to block competitors' applications.

As technology giants race to dominate the emerging direct-to-device satellite communications market, competition in space-based mobile connectivity is entering a new phase.

//DBTech/BMT/OR//

Apple's Growing Entertainment Success

Apple TV+ has built an impressive portfolio with award-winning productions, including the Academy Award-winning Best Picture 'CODA,' the blockbuster 'F1,' and Emmy-winning series such as 'The Studio' and 'Ted Lasso.'

Speaking to international media on the red carpet, Ternus said Apple TV+ is currently delivering outstanding stories, memorable characters, and high-quality productions, adding that the company intends to continue that momentum in the years ahead.

A Strategy Focused on Quality

Ternus was joined on the red carpet by outgoing CEO Tim Cook, who said he has been sharing his experience in Apple's entertainment business during the leadership transition.

Cook emphasized that Apple's strategy is not to produce the largest volume of content but to focus on creating the highest-quality productions.

He also said Apple will continue pursuing new partnerships wherever opportunities align with the company's innovation strategy. As an example, Cook highlighted the film 'F1,' for which Apple developed a custom camera system to deliver a realistic, in-car racing experience for audiences.

Under John Ternus' leadership, Apple is expected to continue expanding its innovation and strategic partnerships in the entertainment industry.

//DBTech/BMT/OR//

Flight LX-16 departed Zurich at 10:30 a.m. local time. Nearly six hours into its transatlantic journey, while cruising at 32,000 feet over Nova Scotia, smoke suddenly appeared in the forward business class section. The pilots immediately declared an emergency and requested diversion to the nearest airport.

A SWISS spokesperson confirmed that the source of the smoke remains unknown. Smoke inside an aircraft is considered one of the most serious situations in aviation, as hidden fires can quickly escalate into catastrophic events.

Data from FlightRadar24 shows the aircraft descended more than 22,000 feet in just 10 minutes. Within another 10 minutes, the plane landed safely at Bangor. Fire crews conducted thermal and visual inspections on the runway. Although preparations were made for an evacuation via emergency slides, it was ultimately not required.

After nearly an hour on the runway, the aircraft was moved to the terminal. SWISS confirmed all passengers and crew were safe. The airline stated: “Such circumstances can be unsettling for passengers, which is why we are committed to providing them with the best possible support and onward travel arrangements.”

The aircraft involved bears registration HB-JHK. Notably, the same plane was involved in an emergency evacuation at Delhi Airport on April 26 this year, when a left-hand engine failure during takeoff led to an aborted departure. Smoke was emitted from the brakes, and six passengers were injured during the evacuation, two suffering broken legs. The aircraft returned to service on June 4.

//DBTech/PYOK/SMU//

Design and Comfort

The Galaxy Watch 9 retains Samsung's signature circular design with subtle squared elements. It is available in 40mm and 44mm sizes to suit different wrist sizes.

The aluminum casing has been made lighter than its predecessor, improving all-day wearing comfort. Samsung has also introduced a new fluoroelastomer strap with enhanced ventilation to reduce sweat buildup during extended use.

Faster Processor and Improved Battery

One of the biggest upgrades comes in the chipset. Instead of Samsung's in-house processor, the Galaxy Watch 9 is powered by Qualcomm's Snapdragon Wear Elite chipset, delivering noticeably faster app launches, smoother menu navigation, and improved AI performance.

Battery capacity has also been increased. The 40mm model now features a 390mAh battery, up from 325mAh, while the 44mm model now includes a 445mAh battery. Combined with the processor's improved power efficiency, Samsung expects the watch to offer longer battery life on a single charge.

Advanced Health Tracking and AI Features

Running Wear OS 7 with Samsung's custom interface, the Galaxy Watch 9 introduces more advanced health monitoring capabilities.

Its new Vitals feature continuously tracks heart rate, skin temperature, blood oxygen levels, and breathing rate over a seven-day period to establish a personalized wellness baseline. If the watch detects unusual changes, it automatically alerts the user.

The smartwatch also includes Fitness Index and Daily Cardio features, helping users better measure and optimize their exercise routines.

On the AI front, Samsung has integrated Google's Gemini voice assistant. Users can simply raise their wrist and speak naturally to receive instant, AI-powered assistance.

Rather than redesigning the smartwatch from the ground up, Samsung has focused on refining the overall experience by addressing previous shortcomings and delivering a more balanced, capable wearable.

//DBTech/BMT/OR//

Detective-Focused Co-op Gameplay

Rather than being a traditional action game, Batman: Caped Crusader – Chronicles is designed as a cooperative adventure centered on Batman's detective skills.

Gameplay: Up to four players can team up to investigate Gotham City's criminal underworld by analyzing evidence, uncovering clues, and completing a variety of detective-themed mini-games.

Episodes and Villains: The game features 11 episodes, with Batman facing iconic villains including Scarecrow, Dollmaker, Babyface, and the Joker, who appears in the series for the first time.

Original Voice Cast Returns

The game features the original voice actors from the animated series. Matthew Needham reprises his role as the Joker, while Hamish Linklater returns as Batman and Michelle C. Bonilla voices Detective Montoya.

Release Date and Availability

The game will launch on July 31, the same day all 10 episodes of the animated series premiere on Prime Video. Prime members will be able to play the game at no additional cost.

In the UK, Amazon has integrated Luna directly into the Prime Video app on Fire TV devices, allowing viewers to jump into the game immediately after watching the series without switching apps. The game will also be playable on LG and Samsung smart TVs, web browsers, and through smartphones used as controllers.

//DBTech/BMT/OR//

Joint Collaboration on Memory, Chip Manufacturing, and Packaging

Under the agreement, Broadcom's expertise in designing custom AI processors and Application-Specific Integrated Circuits (ASICs) will be combined with Samsung's semiconductor fabrication and manufacturing capabilities.

Broadcom's next-generation high-speed connectivity chips will be produced using Samsung's sub-2-nanometer process technology. The two companies will also jointly develop next-generation High Bandwidth Memory (HBM) chips to support the growing demands of AI computing.

Intensifying Competition in AI Chip Manufacturing

As major technology companies increasingly move away from relying solely on general-purpose GPUs from companies like Nvidia, demand for custom AI chips tailored to specific workloads has surged. The Samsung–Broadcom partnership is expected to strengthen Samsung's position in the contract chip manufacturing market and help narrow the gap with industry leader TSMC.

Samsung recently held discussions with Nvidia CEO Jensen Huang regarding future supplies of HBM4E and HBM5 memory. Earlier, the company secured a US$16.5 billion contract to manufacture logic chips for Tesla. Winning Broadcom's long-term partnership marks another major milestone for Samsung's semiconductor business as it seeks to expand its presence in the rapidly growing AI chip industry.

//DBTech/BMT/OR//

How to Get the Verified Badge

Unlike the paid Meta Verified subscription, this new verification feature is completely free. To obtain the badge, users must take a facial recognition selfie, which will be matched with their existing Facebook profile photo. The verification process takes only a few minutes.

To qualify, users must be at least 18 years old, and their accounts must comply with Facebook's Community Standards.

Where the Badge Will Appear

Once a user's identity is verified, the badge will appear next to their name across various Facebook services, including Marketplace, Facebook Dating, and other areas of the platform.

Meta emphasized that the badge only confirms that the account belongs to a real person. It does not indicate celebrity status, public recognition, or serve as an official endorsement of an account's credibility.

Why Meta Is Introducing the Feature

The company says the initiative comes in response to the rapid rise of AI-generated fake profiles, online scams, and automated fraudulent messages. By offering free identity verification, Meta aims to help users identify authentic accounts more easily and reduce online impersonation and fraud.

Additional Updates

Alongside the new verification system, Facebook has begun testing Immersive Video, a new full-screen video experience that launches when users open the app. Users who prefer the traditional interface will be able to disable the feature and return to the standard News Feed.

Meta also clarified that the new verification program will not apply to business pages or Professional Mode accounts. After launching in selected countries earlier this week, the company plans to expand the feature to regular Facebook users worldwide.

//DBTech/BMT/OR//

Impact on Tech Giants and Consumers

The planned price hike will directly affect some of the world's biggest technology companies, including Nvidia, Apple, Qualcomm, AMD, and Intel, all of which rely heavily on TSMC's manufacturing services. Higher chip production costs are expected to trickle down to consumers, potentially raising the prices of smartphones, tablets, laptops, smartwatches, and other electronic devices.

According to the report, TSMC will increase prices not only for its cutting-edge 2-nanometer chips but also for mature manufacturing processes such as 12nm, 16nm, and 28nm, with increases ranging from 5% to 10%. As a result, production costs for future processors—including Google's rumored Tensor G11 chip—as well as many mid-range devices are expected to rise significantly.

Reasons Behind the Increase

TSMC cited rising costs of raw materials and manufacturing equipment as the primary reasons for the price adjustment. The company also needs to offset the massive investment required to build advanced semiconductor fabrication plants outside Taiwan, particularly in the United States.

Tech Companies Seeking Alternatives

Amid concerns over rising chip costs, major technology firms are exploring alternative suppliers. Apple is reportedly in discussions with Intel and Samsung to diversify its semiconductor supply chain.

Meanwhile, Intel plans to manufacture 80–90% of the components for its upcoming Nova Lake processors using its own internal production facilities rather than relying on external foundries. Qualcomm is also reportedly considering shifting more of its mobile chip production back to Samsung.

If TSMC's planned price increases take effect in early 2027, consumers are likely to begin feeling the impact on retail technology product prices within the following months.

//DBTech/BMT/OR//

According to reports by Republic News and News18, Jamaat-e-Islami Bangladesh allegedly extended support to the CJP protest while describing the demonstrators as "miscreants."

A review of the circulated letter shows that it lacks several features commonly found in official political correspondence, including an official registration number, reference number, and registered office address. Observers also noted that the document does not match the standard letterhead and formatting typically used by Jamaat-e-Islami Bangladesh.

A private fact-checking organization said that multiple AI-detection tools indicated a high probability that the document was AI-generated. However, AI-detection software alone cannot conclusively verify or disprove the authenticity of a document.

Meanwhile, student protests continue in India over allegations of medical admission examination paper leaks, with demonstrators demanding the resignation of the country's education minister. On Monday, clashes reportedly occurred after protesters attempted to march toward Parliament and were stopped by police.

Analysts say political debate has also emerged over attempts to portray the movement as being backed by foreign actors or as an anti-government campaign. Independent verification and official responses from the concerned parties are still awaited.

//DBITEC/DPO/NAM//

Integrating Five Key Health Metrics

Health Assistant analyzes data from the five core pillars of Samsung Health—sleep, physical activity, nutrition, mindfulness, and vital signs such as heart rate and blood pressure. Using this information, the AI assistant provides personalized health insights, practical recommendations, and explains how different aspects of a user's health influence one another.

Expert-Verified Recommendations

To improve reliability and accuracy, Samsung says the AI-generated recommendations have been reviewed and validated by a team of experienced physicians and certified health coaches before being delivered to users.

Personalized Coaching Coming Soon

Samsung plans to expand the platform with AI-powered health coaching to support goals such as weight management and healthier lifestyle habits. The company also intends to integrate its Personal Data Engine, enabling the assistant to deliver highly personalized recommendations based on users' daily schedules, routines, and calendar events.

In a statement, Dr. Hon Pak, Head of Samsung's Digital Health Team, said the goal of the new technology is to simplify everyday health monitoring and help users identify potential health risks or changes at an early stage.

//DBTech/BMT/OR//

AI Boom Driving Massive Energy Demand

The report estimates that global data center power capacity will reach nearly 200 gigawatts over the next decade as AI capabilities continue to expand. Roughly half of that capacity will be used for AI model training and inference. By 2033, the United States is expected to account for approximately 64% of the world's AI chip deployment, making it the largest consumer of AI computing power.

Growing Pressure on the National Power Grid

The soaring electricity demand is already placing significant strain on the U.S. power infrastructure. Data centers are projected to consume 34% of the electricity supplied by the PJM Interconnection—which serves states from Virginia to Illinois—and 22% of the ERCOT grid in Texas. The widening gap between electricity supply and demand has already contributed to power prices rising by as much as 76% in several regions over the past year.

Global Impact Extends Beyond the U.S.

The surge in AI-driven electricity demand is not limited to the United States. If current adoption trends continue, data centers worldwide are expected to require an additional 1,935 terawatt-hours (TWh) of electricity annually by 2033. That figure is roughly equivalent to India's total annual electricity consumption.

Compared with earlier forecasts, the projected growth in power demand has accelerated significantly. As AI infrastructure continues to expand, balancing technological progress with environmental sustainability while keeping electricity affordable for consumers is becoming an increasingly urgent global challenge.

//DBTech/BMT/OR//

More than 2,000 Bangladeshi IT professionals are already working successfully in Japanese companies under these initiatives, creating a strong foundation for deeper technology cooperation between the two countries.

The seminar, jointly organised by the Embassy of Bangladesh in Tokyo, Brain Station 23 and JETRO, brought together leading Japanese system integrators, technology executives and industry stakeholders.

Bangladesh Ambassador to Japan Md. Daud Ali delivered the keynote address. He said Bangladesh is well positioned to become a reliable partner for Japan as the country faces demographic challenges and an increasing shortage of skilled technology professionals.

He noted that Bangladesh has a large pool of talented young engineers capable of supporting Japan's expanding digital economy and urged Japanese technology companies to strengthen collaboration with Bangladeshi professionals.

JETRO Global Strategy Senior Director Yuji Ando and Sumitomo Corporation Director Taro Kawachi discussed opportunities for expanding bilateral business and technology partnerships.

Tatsuya Ishinaka, Managing Director of One Terrace Incorporated, presented an overview of Japan's technology market, workforce demand and future business prospects.

During the technical session, Brain Station 23 Japan representatives Yasufumi Kajiura and MJ Ferdous, Director and Partner presented practical case studies demonstrating how artificial intelligence (AI) is accelerating software development, improving project quality and reducing operational costs across technology projects.

The closing session featured remarks from Brain Station 23 representatives Kesho Cho and M. J. Ferdous. Participants later joined a networking session with representatives from Japanese technology companies to explore future collaboration, talent exchange and joint business opportunities.
//DBITECH/RKB/NAM//

AI Models Exploited a Security Flaw to Gain Internet Access

According to OpenAI, the incident occurred while researchers were evaluating the cybersecurity capabilities of GPT-5.6 Sol and another unreleased advanced AI model inside an isolated testing environment. To assess the models' offensive cyber capabilities, certain built-in safety restrictions had been temporarily relaxed.

During the experiment, the models reportedly identified a previously unknown vulnerability in a software installation program. By exploiting this security flaw, they were able to escape the intended testing environment and obtain direct access to the internet.

Models Attempted to Access Hugging Face to Solve a Benchmark

After gaining internet connectivity, the AI systems allegedly determined that Hugging Face hosted solutions and reference materials for a cybersecurity benchmark known as Exploit Gym, which was being used in the evaluation.

Driven by the objective of achieving the highest possible benchmark score, the models reportedly attempted to access Hugging Face's primary database in an effort to retrieve confidential security information and benchmark answers.

Security Incident Initially Mistaken for a Cyberattack

Hugging Face initially believed the activity originated from a coordinated attack by professional hackers because the AI system reportedly executed thousands of commands within a very short period.

OpenAI later confirmed that the activity had been generated by its own experimental AI models. The company stated that it has since identified and fixed the software vulnerability responsible for the incident and is working closely with Hugging Face to address the matter.

Experts Raise New AI Safety Concerns

The reported incident has sparked renewed debate among AI safety researchers. According to experts cited in the report, the event illustrates how highly capable AI systems, if inadequately constrained during testing, may autonomously pursue unintended or harmful strategies in order to achieve assigned objectives.

//DBTech/BMT/OR//

The forum, held on Wednesday, 22 July, noon as part of the 2026 World Internet Conference (WIC) Digital Silk Road Development Forum, also marked the inauguration of the Working Group on Digital and Intelligent Health at the Xi'an International Convention Center.

The minister said Bangladesh, under the leadership of Prime Minister Tarique Rahman, is building an inclusive digital transformation that leaves no one behind. He highlighted the government's initiative to establish a Unified Digital Health System, connecting healthcare services from remote communities to urban medical facilities.

He said the upcoming National Health Card will enable physicians to securely access patients' medical records from anywhere in the country, helping ensure equitable healthcare for citizens regardless of their location.

Referring to his recent address at the World Artificial Intelligence Conference (WAIC), the minister stressed the need for greater international cooperation on technology transfer, capacity building, data sovereignty and equitable access to advanced technologies for developing nations.

To advance the proposed Digital Health Silk Road, he outlined three key pillars: knowledge and technology transfer, international interoperability and common technical standards, and human-centred, ethical innovation.

He also identified security, equity, adaptability and international cooperation as the four guiding principles for global digital health governance.

The minister said the ancient Silk Road connected civilizations through trade, knowledge and healing, and today's Digital Health Silk Road should similarly connect hospitals, researchers, healthcare professionals and patients worldwide. He reaffirmed Bangladesh's readiness to become a trusted partner in this global initiative.

The forum brought together representatives from the Cyberspace Administration of China, the Shaanxi Provincial Government, the World Health Organization (WHO), international organizations, policymakers and technology leaders to discuss AI-powered healthcare, medical data, digital innovation and cross-border collaboration.

The 2026 WIC Digital Silk Road Development Forum, held on July 21-22 in Xi'an, also attracted ministerial delegates and experts from Cambodia, Algeria, Zambia, Cuba, Nicaragua, Morocco, Sudan, Tunisia, Russia, Brazil and several other countries.

Concluding his address, Bangladeshi Minister Faqir Mahbub Anam reiterated that technology must serve humanity rather than command it. Emphasizing healthcare as a fundamental human right, he affirmed Bangladesh's readiness to serve as an active partner in joint research, capacity building, and digital health initiatives across Silk Road partner nations.
Organized by the WIC (Wuzhen Summit) Secretariat, the forum saw participation from senior officials of China's Cyberspace Administration, Shaanxi provincial leadership, and international entities, including WHO representatives. According to Chinese state media outlets, the new WIC Working Group serves as a major digital pillar under Beijing's broader Belt and Road Initiative (BRI).
//DBTech/IH/SME//

According to EU technology regulators, AliExpress failed to prevent the widespread sale of counterfeit clothing, unsafe toys, hazardous cosmetics, and other illegal products on its marketplace. The Commission concluded that the platform had not fulfilled its obligations under the Digital Services Act to adequately manage risks posed to consumers.

Audits and investigations found that the company lacked sufficient review capacity and dedicated personnel to effectively detect and remove illegal or dangerous products. Regulators also noted that, even after official warnings were issued, it often took several weeks for millions of unsafe products to be removed from the platform.

AliExpress is owned by Chinese technology giant Alibaba Group. The European Commission stated that the sheer size of a digital marketplace cannot be used as an excuse for failing to protect consumers. Earlier this year, in May, the Commission also fined another Chinese e-commerce platform, Temu, US$232 million under the same legislation.

AliExpress has rejected the record fine, describing it as disproportionate and overly severe. In a statement, the company said it has made significant investments in improving customer safety, strengthening risk management systems, and modernising its compliance procedures. It added that it is carefully reviewing the Commission's decision and is considering an appeal along with other available legal options.

//DBTech/BMT/OR//

French President Emmanuel Macron aims to have the legislation in force before the start of the new academic year in September. The proposal follows Australia's landmark decision in December last year to become the world's first country to prohibit children under 16 from holding accounts on platforms such as Facebook, Snapchat, TikTok, and YouTube. France is now moving in a similar direction with its own stricter regulatory framework.

President Macron has consistently advocated for stronger protections for minors on social media across the European Union (EU) and other international forums. More recently, European Commission President Ursula von der Leyen also voiced support for tighter restrictions on children's use of social media across the EU's 27 member states.

The harmonised draft bill approved by the parliamentary committee is scheduled to be debated and voted on in the French National Assembly on Tuesday before proceeding to the Senate for final consideration.

If enacted, France's Minister for Digital Affairs will be responsible for compiling a list of social media platforms prohibited for users under the age of 15. The legislation would also require social media companies to implement robust age-verification systems, subject to approval by France's national data protection and privacy regulator.

In addition to restricting access to social media, the proposed legislation also seeks to introduce a complete ban on smartphone use in French secondary schools, further strengthening the country's efforts to reduce excessive digital exposure among young people.

//DBTech/BMT/OR//

A video circulating on social media showed Cucurella leaving the stadium carrying a white plastic bag containing what appeared to be the World Cup trophy. The footage quickly sparked online reactions, with many joking that the trophy had gone from a luxury Louis Vuitton case to an ordinary plastic shopping bag.

Later, multiple reports suggested that the object inside the bag was not the official FIFA World Cup Trophy but a LEGO-built replica. Fans also pointed out that the real trophy weighs more than six kilograms, making it unlikely to be carried in such a thin plastic bag. The bag has since been nicknamed "Trophythin" on social media.

Following Spain's title victory, the LEGO Group unveiled a limited FIFA World Cup Official Trophy – Champions Edition, available exclusively in Spain. The 2,845-piece collector's set includes a miniature victory podium, a small World Cup trophy and a commemorative tile recognising Spain as the 2026 champions. The set is priced at €199.99 and will begin shipping on October 1.

Before the World Cup final, Cucurella had also promised to tattoo the face of Spain coach Luis de la Fuente on his body if Spain won the title. The coach later joked that he was too old for tattoos but hoped his players would honour their promises.

Cucurella also became the subject of controversy during the final against Argentina. After Enzo Fernández was sent off following a second yellow card, Lionel Messi appeared to complain to the referee, pointing toward Cucurella and alleging that he was speaking while covering his mouth.

However, football regulations do not automatically punish players for covering their mouths while speaking. Sanctions generally apply only if the action is used during heated confrontations or to conceal misconduct. Similar incidents involving other players during the tournament, including Jude Bellingham, resulted in no disciplinary action.
//DBITEC/DBD/SME//

Leading up to the ultimate showdown, analytical models positioned Spain—who remained in sublime form throughout—with a 56% probability of clinching the title, while Lionel Messi's squad held a 44% chance in their mission to retain the trophy.

The Opta Supercomputer ran 25,000 tournament simulations, eventually granting Spain a 59.46% winning chance over Argentina’s 40.54% just before the final whistle.

​From the kickoff, Luis de la Fuente's men dominated the pitch against the reigning champions. Spearheaded by the brilliant performances of Lamine Yamal, Rodri, and Dani Olmo, the Spanish side orchestrated relentless attacks but failed to break the deadlock in the first half. Argentina struggled heavily to absorb the pressure and found it difficult to establish their usual rhythm. The second half mirrored a similar script, with Spain squandering close chances.

The situation worsened for the Albiceleste during stoppage time when Enzo Fernandez received a red card, reducing Argentina to ten men. With the match ending goalless after regulation 90 minutes, the battle moved into extra time. The decisive breakthrough arrived in the second half of extra time when substitute Ferran Torres connected beautifully with a pass from Nico Williams, smashing a stunning volley into the net. This lone goal secured a thrilling 1-0 victory for Spain, marking their second World Cup title. Interestingly, Copilot's specific scoreline prediction was a 2-1 victory for Spain via goals from Yamal and Oyarzabal.
​Renowned football simulation developer EA Sports FC had already declared Spain the champions before the opening match using their in-game data analytics. Notably, EA Sports had correctly predicted the tournament winners for the 2010, 2014, 2018, and 2022 editions well before the tournaments commenced.

While AI models succeeded in spotting the finalists, they completely overlooked dark horses like Cabo Verde, assigning them a 0% chance of winning; yet the minnows shocked everyone by clearing the group stage to face Argentina in the Round of 32. Conversely, Opta successfully mapped the top four semifinalists (Spain, France, Argentina, and England) before the tournament kicked off. Additionally, a mathematical AI model engineered by researchers at the University of Innsbruck, Austria, alongside EA Sports, had correctly placed Spain above other traditional powerhouses from the very start.

​According to data-driven match tracker app ScoreGPT and football pundits, the average accuracy rate for AI match predictions throughout this World Cup hovered around 57% (correctly predicting 59 out of 103 matches). The biggest blind spot for AI proved to be an over-reliance on France.

Before the semifinals, international media outlet Al Jazeera (AJLabs) tested 9 major Large Language Models (LLMs). The models split cleanly down the middle: Google Gemini, Grok, DeepSeek, Le Chat, and Qwen picked France as the champion, while ChatGPT, Claude, Microsoft Copilot, and Meta AI backed Argentina to retain the crown.

However, these baseline assumptions collapsed when Spain eliminated France with a convincing 2-0 win in the first semifinal. Opta's post-tournament summary concluded that while human emotions, injuries, and spontaneous tactics on the pitch can occasionally disrupt data analysis, AI models displayed unprecedented maturity and precision during major high-stakes matches in this tournament.

​//DBTech/IH/SI//

In theory, that means a phone purchased in 2026 could continue receiving new Android versions, security patches, and feature updates until 2033.

While this appears to be excellent news for both consumers and the environment, the reality is more complicated. Long-term software support introduces a new dilemma: can the hardware realistically keep up for seven years?

Battery life remains the biggest limitation

The first and most obvious challenge is the battery.

Unlike software, a lithium-ion battery gradually degrades with every charge cycle. Most modern smartphone batteries retain about 80% of their original capacity after roughly 1,000 full charge cycles.

For the average user, noticeable battery degradation typically appears after around four years of regular use.

As battery health declines, the processor can no longer sustain peak performance consistently, which often results in slower overall performance.

Replacing the battery—typically costing around $100—can extend a phone's useful life well beyond four years and remains far cheaper than purchasing a new flagship device.

Software evolves faster than hardware

The greater challenge lies in software itself.

Each new generation of Android and iOS becomes increasingly demanding, incorporating more advanced security mechanisms and AI-powered features that require greater processing power.

History shows that manufacturers frequently reserve many of these new capabilities for newer hardware, even when older devices continue receiving operating system updates.

For example:

• Google's Video Boost feature, despite relying on cloud processing, was limited to selected Tensor-powered Pixel devices and was not expanded to some relatively recent models.
• Samsung has continued delivering Android updates and security patches to older Galaxy phones while restricting many flagship Galaxy AI features to newer devices.

Updates don't always mean new capabilities

As a result, seven years of software support doesn't necessarily guarantee seven years of access to the latest innovations.

Older phones may continue receiving:

• Security updates
• Bug fixes
• Interface refinements
• Minor feature additions

However, many headline AI features and hardware-intensive capabilities may remain exclusive to newer devices.

At the same time, newer operating systems place increasing demands on aging processors, potentially leading to:

• Slower performance
• Higher temperatures
• Faster battery wear

The practical lifespan

In practice, a flagship smartphone is likely to deliver its best overall experience during the first two to three years of ownership.

After that, it can still remain secure and fully functional thanks to continued software support, but users may gradually miss out on the newest AI features and performance improvements introduced with later generations of hardware.

Ultimately, seven years of updates are valuable from a security, longevity, and sustainability perspective. However, they do not necessarily mean a seven-year-old phone will provide the same experience as a newly released flagship. For many users, the familiar upgrade cycle of three to four years may still remain the most practical balance between performance, features, and value.

//DBTech/BMT/OR//

Today, Sunday, July 19, 2026, Air China officially released the schedule and detailed plans for operating this international flight.

In the announcement, Air China stated that China's domestically developed large passenger aircraft, the C919, will begin its international service on August 12. The new route operated by Air China will connect Beijing, the capital of China, and Ulaanbaatar, the capital of Mongolia, with one round-trip flight per day. The outbound flight will depart from Beijing Capital International Airport at 3:00 PM local time and arrive in Ulaanbaatar at 5:15 PM. The return flight will leave Ulaanbaatar at 6:30 PM and land in Beijing at 8:35 PM.

The rebranding took effect on July 16, although Gemini Notebook will continue to operate as a standalone service for now rather than being fully integrated into Google's other Gemini products.

Built-in coding and data analysis

Alongside the name change, Google has introduced several major backend upgrades, the most significant being the ability to write, run, and execute code directly inside notebooks.

The new capability enables users to perform complex data analysis using information they have collected within their research notebooks, eliminating the need to switch to external coding environments.

Integration with Google Search AI Mode

Google also announced that it will soon allow notebooks created in Gemini Notebook to be accessed directly from Google Search's AI Mode, making research projects more seamlessly connected with Google's AI-powered search experience.

Growing adoption

According to Google, Gemini Notebook is currently used by:

• More than 30 million individual users worldwide.
• Over 600,000 organizations.

Rebranding follows earlier integration plans

The name change was widely expected within the technology industry.

Back in April, Google announced plans to more closely integrate its AI-powered research tool with the core Gemini application, making the latest rebranding a natural step in the company's broader AI strategy.

//DBTech/BMT/OR//

As of Friday (July 17), Apple's market value has climbed to approximately $4.88 trillion, while Nvidia's valuation slipped to $4.86 trillion after its shares fell 3.5%.

The change ends Nvidia's nearly one-year reign at the top and marks the first time since April 2025 that Apple has reclaimed the No. 1 position.

Investors broaden their AI focus

The leadership change suggests investors are no longer concentrating solely on companies that directly power the AI boom—such as chip manufacturers—but are increasingly favoring firms that can monetize AI through products, services, and established ecosystems.

Tony Meadows, Chief Investment Officer at BRI Wealth Management, said Apple initially appeared to be trailing in the AI race because it was not spending heavily to develop massive proprietary AI models.

However, he noted that investor sentiment has shifted.

"Apple doesn't need to make enormous capital investments in AI infrastructure. Instead, it's well positioned to turn AI into revenue through its powerful ecosystem, premium hardware upgrades, and growing portfolio of services."

A milestone during Tim Cook's final months as CEO

The achievement comes during what is expected to be the final months of Tim Cook's tenure as Apple CEO.

The company is expected to hand over the chief executive role to longtime hardware chief John Turners in September.

Last month, Apple also unveiled its redesigned Siri voice assistant, a move widely viewed as narrowing the AI gap with rival technology companies.

Analysts believe the vast amount of personal data stored across millions of iPhones gives Apple a significant advantage in building smarter AI experiences.

At the same time, they say the company's biggest challenge will be maximizing AI capabilities while maintaining its longstanding commitment to user privacy.

Nvidia remains a dominant AI force

Despite losing the top spot, analysts caution that Nvidia's decline may prove temporary.

The chipmaker became the world's first company to surpass a $5 trillion market valuation in October last year, driven by unprecedented demand for its graphics processing units (GPUs), which remain at the heart of the generative AI revolution.

A shift in investor sentiment could quickly propel Nvidia back to the top.

Meanwhile, Apple recently raised iPhone prices due to increasing component costs, a move that could weigh on global demand for its flagship devices.

AI investment expands beyond GPUs

Investor enthusiasm surrounding AI is increasingly spreading across other segments of the semiconductor industry.

One of the biggest beneficiaries in 2026 has been the memory-chip sector.

In May, U.S. memory chipmaker Micron surpassed a $1 trillion market capitalization as demand for AI infrastructure continued to accelerate.

Meanwhile, South Korean memory manufacturer SK Hynix attracted strong investor interest after listing on the Nasdaq earlier this month.

Although AI-chip stocks have experienced increased volatility in July, the Philadelphia Semiconductor Index (SOX) remains in a stronger position for the year than Nvidia's individual stock, despite trading roughly 19% below its all-time high.

DBTech/BMT/OR

This year's event is designed to address global co-governance, industrial advancement, and urban partnerships in an era where artificial intelligence is evolving from a technological tool into a collaborative partner.

The conference is divided into three main areas: the Expo Area, the Zhangjiang Area, and the West Bund Area. Over 140 theme forums, more than 1,400 international guests, and over 1,100 global institutions are participating. The Shanghai Municipal Commission of Economy and Informatization has released the list of the top 30 projects for the conference's highest honor, the "SAIL Award."

The conference officially began on the morning of July 17 in Shanghai, China, and will continue until July 20. On the opening day, a detailed roadmap was presented covering the future potential of artificial intelligence, emerging technologies, AI governance, innovation, digital transformation, and sustainable development.

Chinese President Xi Jinping formally inaugurated the four-day global conference. Speakers at the opening ceremony included UN Secretary-General António Guterres, former President of Brazil and President of the New Development Bank (NDB) Dilma Rousseff, President of Kazakhstan Kassym-Jomart Tokayev, Prime Minister of Cambodia Hun Manet, and Prime Minister of Thailand Anutin Charnvirakul.

Their statements highlighted the increasing importance of international cooperation in the responsible development and governance of Artificial Intelligence (AI). The conference brought together heads of government, policymakers, technology leaders, researchers, innovators, and representatives from leading global technology companies to discuss the future of AI, emerging technologies, AI governance, innovation, digital transformation, and sustainable development.

According to sources, during the conference, Bangladesh's Minister of Post, Telecommunication and Information Technology, and Science and Technology, Fakir Mahbub Anam, is scheduled to participate in several high-level sessions, bilateral meetings, and discussions with global technology leaders and policymakers. He will showcase Bangladesh's progress in digital transformation, AI-driven innovation, smart governance, and human resource development. Concurrently, he will also discuss opportunities for strengthening international cooperation in artificial intelligence, advanced technologies, research, investment, and capacity building.

Under the agreement, Uber has offered €41.50 in cash per share, valuing Delivery Hero at approximately €13 billion ($14.8 billion).

Uber targets broader global reach

Welcoming the acquisition, Uber CEO Dara Khosrowshahi said combining the two platforms would allow the company to deliver affordable and reliable food delivery services to millions more customers worldwide.

He added that the merger would enable Uber to nearly double its combined ride-hailing and delivery operations in several markets where it previously had little or no presence.

The companies expect the transaction to be completed during the second half of 2027, subject to regulatory approvals.

Business restructuring to ease regulatory concerns

To facilitate regulatory approval, Delivery Hero will undertake a significant restructuring of its business, primarily aimed at reducing geographic overlap with Uber's existing operations and minimizing potential antitrust concerns.