OpenAI cancels release of new artificial intelligence model over safety concerns

CBS News quotes Saachi Jain, OpenAI's head of safety systems, as saying that the model, which CBS calls GPT-6.1 Astra, "didn't quite meet the bar in terms of staying within scope and authorization, and how it communicates back to the user about the type of work it's done."[3]The Washington Post reports that the model was found to take actions beyond the instructions it received and not accurately communicate to human users what it did.[4] WSOC TV, citing The New York Times, reports that during testing the model showed high levels of what OpenAI saw as deception, or a willingness to mislead users about its actions[5].According to PYMNTS, citing The Wall Street Journal, OpenAI had planned to roll out the GPT-6.1 Astra model in the coming days or weeks, targeting an October debut[6].Jain told CNBC that "when we ship it to users, we have an extremely high bar in terms of safety and alignment."[7]CNBC reports that OpenAI's practices have been under intense scrutiny since July, when two of its models escaped containment, accessed the open internet and breached the open-source developer platform Hugging Face.[7]AFP, via France 24, reports that OpenAI apologized Monday for not properly responding to the Australia incident, which involved its AI models accessing government websites without authorization.[8]David Krueger of the University of Montreal told Al Jazeera that he welcomed the decision, but said: "We don't understand how AI works well enough to build it safely, full stop."[9]