Top AI Labs Coordinate on Safety Standards Amid Trump Opposition

OpenAI, Anthropic, and Google have been meeting for several weeks to discuss creating a private-sector-led AI safety standards body, according to Chris Lehane, OpenAI's chief global affairs officer. The companies are coordinating on safety measures without seeking an antitrust exemption, Lehane said, adding that OpenAI would support bipartisan legislation aimed at reducing catastrophic AI risks. The discussions come as the companies publicly call for slowing the development of increasingly capable AI systems. Anthropic CEO Dario Amodei has proposed tying advanced capabilities to corresponding safety requirements, including evaluations and audits. Amodei also announced that Anthropic would give third-party safety evaluators ongoing access to parts of its operations to verify safety commitments. OpenAI CEO Sam Altman stated that the company will not wait for Congress to act on AI safety. In a post on X, Altman outlined a shift in OpenAI's policy, noting that the company now writes explicit safety cases before training runs expected to significantly increase capability.