TL;DR
European leaders used a June 17 G7 working lunch in Évian-les-Bains to press top AI executives on access to frontier models after a U.S. export-control directive forced Anthropic to cut off two models worldwide. The talks exposed a gap between what Europe wants, including durable access and less reliance on U.S.-controlled systems, and what company leaders can promise while Washington controls export policy.
European leaders pressed the heads of Anthropic, Google DeepMind and OpenAI at a June 17 G7 working lunch in Évian-les-Bains, France, after a U.S. export-control directive forced Anthropic to block access to two of its most capable models worldwide, raising direct questions about Europe’s reliance on AI systems controlled under U.S. policy.
The meeting brought Dario Amodei of Anthropic, Demis Hassabis of Google DeepMind and Sam Altman of OpenAI to the same table as heads of state and senior officials. According to the source material, French President Emmanuel Macron devoted the lunch to artificial intelligence, with European leaders including German Chancellor Friedrich Merz, European Commission President Ursula von der Leyen and U.K. Prime Minister Keir Starmer taking part.
The immediate trigger was a June 12 U.S. Commerce Department export-control directive that ordered Anthropic to block its Fable 5 and Mythos 5 models for any “foreign national,” according to the source material. Because nationality could not be reliably checked at API scale, Anthropic’s practical response was a worldwide shutdown of those models. European companies and public institutions that had built systems around them lost access with no warning period, according to the same account.
The AI executives offered versions of international coordination rather than unilateral company control. Amodei called for a U.S.-led coalition of democratic states with structured access for trusted partners, while Hassabis backed a Western coalition. Altman proposed an international forum for model testing standards and said no single lab should make such decisions alone, according to the source material.
Évian and the fallout: what Europe actually wants
For the first time, Amodei, Hassabis, and Altman sat with heads of state — five days after Washington switched Anthropic’s models off worldwide. Europe’s question: can you rely on models a foreign cabinet can shut down by decree?
The dilemma: what Europe wants from the three CEOs, the three can’t deliver — because they don’t hold the switch, Washington does. Macron’s platform is the right answer, but no fix for a decade-old infrastructure gap. The only answer that doesn’t depend on someone else’s goodwill: your own models, your own compute, open weights you can self-host.
Europe Seeks AI Access Guarantees
The shutdown turned a policy concern into an operational problem. For European businesses and public agencies using U.S.-built AI models, the issue is no longer only price, performance or safety. It is whether access can disappear because of a foreign government order.
Europe’s demands, as described in the source material, center on reliable access to frontier models, protection against another abrupt shutdown, a trusted-partner framework for non-U.S. allies and a greater say over where compute infrastructure, chips and power capacity are built. Those demands go beyond what AI companies can decide on their own, since export controls are set by governments.
The episode also adds pressure to Europe’s push for technological sovereignty. The source material cites a €420 billion package, AI gigafactories and the CADA initiative as part of Europe’s response. The policy goal is to reduce dependence on systems that can be limited by outside authorities while still keeping access to leading AI tools.
AI model access control API
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
A Shutdown Changed The Agenda
The G7 meeting had an official theme of safe, rapid and effective AI deployment, according to the source material. But the Anthropic shutdown shifted the discussion toward control, access and geopolitical leverage.
The guest list reflected that wider debate. Alongside U.S. executives, other technology leaders attended, including Marc Benioff of Salesforce and Meta’s AI chief Alexandr Wang. European and allied AI firms were also represented, including Mistral from France, Synthesia from the U.K., Black Forest Labs from Germany, Domyn from Italy and Sakana AI from Japan.
The source material frames the meeting as the first time Amodei, Hassabis and Altman sat with heads of state in this setting. That symbolism mattered because the discussion was not only about product rules or safety testing. It was about whether the infrastructure behind advanced AI will be treated as a shared democratic asset, a private company service or a tool subject to national power.
enterprise AI model management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Access Rules Remain Unsettled
It is not yet clear whether the U.S. directive will be revised, whether Anthropic’s blocked models will return to European users, or whether trusted-partner access can be made legally durable. The source material says the ban remained in place after the summit.
It is also unclear how much the AI companies can offer without a policy change in Washington. The main gap exposed at Évian is that European leaders want guarantees from executives who do not control the government switch that can restrict access.
AI model security tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
September Talks May Set Terms
The source material says leaders are expected to reconvene in September, after a platform for Western democracies is developed within one month. That process is expected to cover trusted-partner access, cyber-defense cooperation, China-related controls and child safety principles.
Europe’s parallel track will be its own infrastructure buildout. The practical test will be whether European governments can turn funding, compute projects and domestic AI firms into systems that public agencies and companies can run without relying on access that another government can withdraw.
international AI testing standards
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What happened at the G7 meeting in Évian?
European leaders met leading AI executives, including Dario Amodei, Demis Hassabis and Sam Altman, to discuss AI deployment and access after a U.S. export-control order forced Anthropic to shut off two major models worldwide.
Why did Anthropic shut down Fable 5 and Mythos 5 worldwide?
According to the source material, a U.S. Commerce Department directive ordered Anthropic to block access for any “foreign national.” Because nationality could not be checked reliably in real time at API scale, the company’s practical option was a worldwide shutdown.
What does Europe want from U.S. AI companies?
European leaders want reliable model access, protection against sudden shutdowns, a trusted-partner access system, a say in AI infrastructure decisions and stronger safety rules, including protections for children and young people.
Can Amodei, Hassabis and Altman guarantee Europe access?
Only partly. The companies can support access frameworks and standards, but export controls are government decisions. That means any durable guarantee would likely require U.S. policy changes or formal agreements between governments.
What is expected after the summit?
According to the source material, Western democratic leaders are expected to work on a shared platform within a month and meet again in September. The Anthropic restriction remained in place after the summit.
Source: Thorsten Meyer AI
