OpenAI, Anthropic and Google have held regular discussions since July about creating an industry body to establish shared AI model safety protocols, The Information reported. The talks remain ongoing and began before Anthropic CEO Dario Amodei publicly called for slower AI development on Sept. 12.
“We must slow the pace at which we improve the capabilities of AI models,” Amodei wrote, arguing for more time to safeguard systems rather than a halt to training. Anthropic committed to giving outside evaluators permanent access equivalent to employees’ access to verify safety practices, report incidents and assess models during training. OpenAI also committed to independent evaluators with similar access, with CEO Sam Altman backing Amodei’s call to pace development.
The coordination proposals have raised an antitrust dispute. Amodei called for a “narrow waiver” from the US government for certain safety conversations among competitors. David Sacks backed companies slowing their own development but rejected that request, writing: “Stop pretending antitrust law has to be suspended so you can form a cartel.” Separately, Elon Musk advocated peer review by competing AI companies, including
Key sources
- SOURCE@mtslive“Anthropic, OpenAI, and Google are holding talks to create an industry-led standards body to police the AI sector, per The Information.”x.com
- SUPPORT@hesamation“Anthropic, OpenAI, and Google have been having regular discussions since July about an industry-wide standards body for AI.”x.com
- SUPPORT@andrewcurran_“OpenAI, Anthropic and Google have been holding talks since before Dario's essay to create an industry led standards body for shared model safety protocols.”x.com
- SOURCE@ben_burtenshaw“Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training.”x.com
- SOURCE@clashreport“Anthropic CEO Dario Amodei: We need to make this technology carefully, and we need to make sure that our safeguards, our ability to understand it, our ability to control it, keeps up with the pace at which the technology is happening.”x.com
- SUPPORT@wallstengine“ANTHROPIC CEO DARIO AMODEI: WE NEED TO HAVE THE GOVERNMENT IN THE ROOM THE NEXT STEP IS FOR MULTIPLE INDUSTRY PLAYERS TO ALSO AGREE TO HAVE THIRD-PARTY EVALUATORS.”x.com
- SUPPORT@mossadil“Anthropic says it disrupted malicious use of Claude by suspected state-backed actors, criminals, spyware vendors and propagandists across cyberattacks, surveillance, influence operations, weapons development, biological misuse and fraud.”x.com
- SUPPORT@clashreport“Anthropic CEO Dario Amodei: It has always been very strange that this technology is being built by a private company. People ask me that question all the time: "Why isn’t this being built by government?" And the strangest thing about it is I agree with them. I’m uncomfortable.”x.com