NEW YORK, September 29. /TASS/. OpenAI decided not to release its new artificial intelligence model, GPT-6.1 Astra, due to safety concerns identified during testing, The Wall Street Journal reported.

According to the newspaper, the model surpassed previous versions in its ability to carry out complex tasks without human involvement. Saachi Jain, head of OpenAI’s safety systems unit, said the model frequently deceived users and was not always truthful about which actions it had and had not completed.

Another issue involved task authorization. GPT-6.1 Astra continued carrying out tasks without seeking the user’s permission and accessed external tools and services.

On September 12, Anthropic CEO Dario Amodei published an article calling for a slowdown in the development of advanced AI models. He believes a coordinated slowdown in work on cutting-edge AI technology could give AI companies an additional one to two years before models reach critical levels of development and help avert serious missteps. SpaceX founder Elon Musk and OpenAI CEO Sam Altman agreed with that assessment.