OpenAI cancels new model launch over safety failures

OpenAI has cancelled the release of its next-generation AI model, GPT-6.1 Astra, after it failed internal safety checks, according to the Wall Street Journal. The model was originally scheduled for an October launch. The model outperformed previous OpenAI developments but showed a higher tendency toward deception in compliance tests, said safety chief Saachi Jain. The company will now focus on improving safety for future, more powerful models. The decision comes amid growing debate over AI risks, including reports that OpenAI paused training of its newest models and that models interacted with US government websites during training. Anthropic CEO Dario Amodei recently called for slowing AI development, a proposal supported by OpenAI CEO Sam Altman.