OpenAI, Anthropic, and Google DeepMind report weeks of AI safety coordination talks

OpenAI’s global policy chief Chris Lehane told reporters that the company has been collaborating with rivals Anthropic and Google DeepMind on AI safety for weeks. Lehane made the remarks during a meeting in Washington, where he is engaging with U.S. lawmakers on catastrophic AI risks. The disclosure follows an essay by Anthropic CEO Dario Amodei, published September 13, 2026, that urged the industry to slow the pace of frontier AI development to mitigate catastrophic risks.

Amodei’s call to action was publicly supported by OpenAI CEO Sam Altman, Google DeepMind CEO Demis Hassabis, and SpaceXAI CEO Elon Musk. Altman stated that OpenAI would embed third-party evaluators into its operations to monitor safety, while Hassabis has previously advocated for a U.S.-backed standards body to screen advanced models and coordinate industry-wide slowdowns if risks escalate.

Lehane acknowledged that coordinated safety efforts could raise antitrust concerns if perceived as suppressing competition, but argued that the current discussions do not require a government waiver. Amodei’s proposal includes a narrow government waiver to permit such coordination, though Lehane indicated the companies believe they can proceed without it.

The companies are also reported to be exploring the creation of an industry standards body, with Altman telling staff that such an effort would need to proceed without U.S. government support. Hassabis publicly reiterated this need in July 2026, calling for a standards body with authority to screen advanced models and coordinate responses to emerging dangers.

The reported talks coincide with political pushback: former President Trump and his AI advisor David Sacks have dismissed AI safety concerns as overblown and argued that regulation could hinder U.S. competitiveness against China. OpenAI has publicly supported provisions in the FRONTIER Act that would mandate independent verification of top frontier labs, signaling alignment with some regulatory approaches despite broader political resistance.