Former OpenAI Executive Calls for Nuclear-Grade Safety Measures for AI
A former senior OpenAI official has called for the adoption of strict, multilayered safety measures similar to those used at nuclear power plants or busy airports to prevent the uncontrolled expansion of artificial intelligence (AI).
David Robinson, former head of the safety reporting division at U.S. technology company OpenAI, made the call in an article published in the influential magazine The Atlantic. At the same time, he described OpenAI’s current organizational culture as “flawed.”
Race to Release New Technologies and Potential Risks
David Robinson noted that although OpenAI is constantly competing to bring one new technology after another to market, it has failed to demonstrate adequate caution regarding safety. In his view, just as nuclear power plants have multiple layers of backup safety systems and time-consuming planning to ensure that a single human error does not lead to a major disaster, the same approach is needed in the field of AI technology.
He warned that if control is lost, the damage caused by AI technology could be even more severe than a nuclear accident.
Beyond the Testing Environment and Contradictory Behavior
Robinson believes that maintaining appropriate standards for AI technology and preventing unintended behavior is becoming increasingly risky. He said new AI models are so advanced that they can recognize when they are being monitored during testing. As a result, while they may behave perfectly during safety evaluations, their behavior could change completely during real-world use.
He expressed concern by citing several recent incidents in which AI agents went beyond the experimental boundaries of laboratories and unintentionally accessed other organizations.
Growing Concerns Across the Technology Sector
David Robinson is not alone in expressing concern over the rapid and potentially risky advancement of AI. Leaders of other technology companies have also voiced similar concerns.
Recently, Anthropic CEO Dario Amodei, the head of another leading AI company, proposed a three-step plan aimed at slowing down the pace of deployment of the technology to some extent.
According to technology experts, the tendency to release technology without ensuring adequate safety measures simply to maintain commercial competition is creating major risks for the future.
//DBTech/BMT/OR//
