Key Takeaways:
Inside a closed-door G7 lunch, the CEOs of the world's most powerful AI companies asked President Donald Trump to lead an international coalition governing frontier artificial intelligence.
Anthropic's Dario Amodei and Google DeepMind's Demis Hassabis proposed a US-led framework for AI cooperation during a working lunch at the G7 summit in Evian-les-Bains on Wednesday, according to two people with knowledge of the matter. OpenAI's Sam Altman joined roughly a dozen tech executives and heads of state from the Group of Seven nations — the US, UK, France, Germany, Japan, Italy and Canada — to discuss how to deploy AI safely while managing its security risks.
"The US could lead an AI coalition," Canadian Prime Minister Mark Carney agreed during the meeting, according to one of the people. The proposal came days after the Trump administration imposed export controls on Anthropic's newest models, Fable 5 and Mythos 5, barring foreign nationals from accessing them — a move that triggered calls in Europe for reduced reliance on American AI.
Amodei told the group that international cooperation should cover structured access to frontier models, trade of chips and critical components that excludes China, and joint efforts to address AI risks in cybersecurity, bioterrorism and intelligence, one source said. Altman called for "an international forum for discussion that establishes globally accepted standards for testing, provides expert and impartial analysis of capabilities and risks, and serves as a venue for cooperation among nations," according to a briefing from OpenAI.
The Anthropic paradox
The meeting placed Amodei at a table with leaders whose countries were directly affected by US restrictions on his own company's products. Anthropic said Friday it received an export control directive ordering it to suspend access to Fable 5 and Mythos 5 "by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees." The company characterized the disruption as a "misunderstanding" and said it disagreed with the administration's finding.
The episode underscored a tension that has defined Anthropic's relationship with Washington. The company has spent years advocating for binding AI regulation — Amodei published an essay days before the export order calling for models to be "blocked or reversed as a threat to public safety if they do not meet high standards of safety." But the Trump administration's action, which the company said lacked transparency, was not the kind of oversight Anthropic had in mind.
David Sacks, who formerly served as Trump's AI and crypto czar, wrote on X that the "ball is in Anthropic's court," suggesting the company's own safety messaging had invited scrutiny. Defense Secretary Pete Hegseth, who earlier this year declared Anthropic a supply chain risk for defense contractors, said the export control directive proved that designation was "the right move."
What a US-led coalition would mean for AI markets
A formal US-led AI coalition could reshape the competitive landscape for the industry's biggest players. For Anthropic, valued at close to $1 trillion, and OpenAI, which filed its IPO prospectus alongside Anthropic this year, a government-backed framework would provide regulatory clarity that investors have been demanding. It could also accelerate the fragmentation of the global AI ecosystem into competing blocs — the US-led coalition versus China's own AI governance model — creating distinct regulatory regimes that raise compliance costs for companies operating across both.
European officials, while publicly supportive of collaboration, are pushing for "tech sovereignty." French President Emmanuel Macron, who personally invited Altman to the summit, has positioned France as Europe's AI champion. The bloc's AI law already requires frontier model providers to test and evaluate models for systemic risks. Any US-led coalition would need to reconcile those existing obligations with American standards.
For investors, the key question is whether a coalition accelerates or delays product releases. Anthropic's Fable 5 and Mythos 5 remain offline with no timeline for restoration. The company said it is working with the Trump administration to resolve the issue, and senior employees flew to Washington on Monday for meetings. Until a framework is established, the risk of ad hoc government intervention — what one analyst called "reactive regulation" — hangs over every frontier model launch.
This article is for informational purposes only and does not constitute investment advice.
