Google, OpenAI and Anthropic are moving closer to creating a proposed independent organization for frontier artificial intelligence standards, provisionally called the Standards Authority for Frontier AI, or SAFA. The Information reported that the body would address gaps in government oversight as debate continues over how much regulation increasingly capable AI systems require. The initiative comes amid opposition from US President Donald Trump to additional AI guardrails and wider disagreement over whether existing laws and market incentives are sufficient. OpenAI and Anthropic have also reportedly discussed allowing each company to test the other’s commercially available models for vulnerabilities and unexpected behavior, while limiting retention of evaluation data. OpenAI chief executive Sam Altman and Anthropic chief executive Dario Amodei recently addressed the UN Security Council on AI safety. Altman said major AI decisions should involve democratic institutions and governments, while Amodei called for international agreements and global testing standards. OpenAI has separately urged the US to lead international technical standards for advanced AI, including systems capable of recursive self-improvement. The proposed industry body reflects efforts by leading developers to establish shared safety practices while governments continue to debate the appropriate scope of oversight.
