OpenAI, Anthropic and Google say they have spent weeks talking AI safety. Will the three announce a joint standards body by Oct 12?
57% of users predicted YES — the community missed this one. 14 predictions cast.
OpenAI, Anthropic and Google had not announced a joint AI safety standards body by October 12, 2026. Reporting through September and into October described the three companies as still negotiating the shape of a voluntary, industry-led body rather than preparing to formally launch one. One proposed name circulating in coverage was the "Standards Authority for Frontier AI," though the companies involved had not confirmed a final structure, and reports disagreed on whether the body would include government oversight or operate as industry self-regulation.
The talks trace back to a July proposal from Google DeepMind co-founder Demis Hassabis for a U.S.-led standards body modeled partly on the financial industry's FINRA. OpenAI policy chief Chris Lehane confirmed the three companies were collaborating and argued an antitrust waiver should not be necessary for safety-focused coordination, but unresolved antitrust questions were among the reasons cited for the delay. Separately, Anthropic CEO Dario Amodei published an essay in mid-September calling for slower capability development and proposing independent evaluators with employee-level model access, reflecting continued internal debate over governance approaches even as a formal joint body remained unannounced.
The Predict Six community got this one wrong, with 57% predicting an announcement that did not materialize by the deadline.