OpenAI Joins Anthropic and Google DeepMind in Push for Coordinated AI Safety Standards

OpenAI has moved into several weeks of coordinated AI-safety discussions with Anthropic and Google DeepMind, centering on an independent standards body and a broader effort to slow frontier-model progress long enough for safeguards to catch up. The effort includes talks over an independent safety standards body, as Anthropic, OpenAI and Google DeepMind leaders converge on a more cautious approach to increasingly agentic systems.