OpenAI will not launch its new GPT-6.1 Astra AI model due to security concerns, the maker of ChatGPT confirmed on Tuesday. The AI system, which performs tasks like browsing the web and using apps on its own, “didn’t meet the bar” of the company’s standards, said Saachi Jain, head of security systems at OpenAI. On Tuesday, OpenAI also published an update on incidents that occurred in June but were not made public until last week, in which its models accessed Australian government websites. and unauthorized systems. Such incidents and similar breaches by models developed by leading AI companies have in recent weeks intensified debate over the risks posed by the technology. Top AI leaders, including OpenAI’s Sam Altman and Anthropic chief Dario Amodei, have urged the industry to slow the pace of development due to concerns about the risks associated with the technology. “We want to make sure that our model development is secure, regardless of whether it’s in the enterprise or when we ship it to users. But when we ship it to users, we have an extremely high bar in terms of security and alignment,” he added. The GPT-6 Astra flagship agent model was launched in September and specializes in complex reasoning and execution tasks. autonomously. OpenAI said it was the result of “years of research and big bets.” OpenAI will hold its annual DevDay developer conference in San Francisco on Tuesday, where it is expected to make several announcements. It is unclear whether a new version of Astra will be among them. The company’s security controls have come under intense scrutiny after several high-profile incidents involving its technology.
