Meta Launches AI Agent to Automate Smartphone Tasks

Technology giant Meta has launched a new artificial intelligence (AI) agent designed to automatically perform complex tasks within various smartphone apps, such as sending emails, booking travel tickets and purchasing products on behalf of users. Named “Muse,” the AI assistant has been described by Meta CEO Mark Zuckerberg as a key tool for building “personal superintelligence.”

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//