TL;DR
European leaders used a June 17 G7 working lunch in Évian-les-Bains to press leading AI executives on access, safety and dependence after a U.S. export-control directive disrupted Anthropic model access. The core issue is whether Europe can rely on frontier AI systems controlled by companies subject to U.S. government orders.
European leaders met Dario Amodei, Demis Hassabis and Sam Altman at a June 17 G7 working lunch in Évian-les-Bains, France, after a U.S. export-control directive pushed Anthropic to block its most capable models worldwide, turning Europe’s concern about AI dependence into an immediate policy and business risk.
The source material says the June 12 U.S. Commerce Department directive ordered Anthropic to block access to its Fable 5 and Mythos 5 models for any “foreign national.” Because nationality cannot be reliably checked in real time at API scale, Anthropic’s response was a worldwide shutdown, according to the account. European companies and public institutions that had built those models into operations lost access without lead time.
Five days later, French President Emmanuel Macron devoted a G7 working lunch to artificial intelligence. Amodei of Anthropic, Hassabis of Google DeepMind and Altman of OpenAI sat with heads of state and senior officials, including Macron, German Chancellor Friedrich Merz, European Commission President Ursula von der Leyen and U.K. Prime Minister Keir Starmer. President Donald Trump attended with Treasury Secretary Scott Bessent, Commerce Secretary Howard Lutnick and Secretary of State Marco Rubio, according to the source material.
The official framing was safe, rapid and effective AI deployment. The sharper dispute was whether allies can depend on American AI companies when access to their systems can be shaped by the U.S. executive branch. The account says other technology leaders and allied AI labs were also present, including Salesforce’s Marc Benioff, Meta AI chief Alexandr Wang, France’s Mistral, the U.K.’s Synthesia, Germany’s Black Forest Labs, Italy’s Domyn and Japan’s Sakana AI.
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.
Model Access Becomes Policy Risk
The Évian meeting matters because frontier AI is no longer only a product decision for companies buying software. For banks, manufacturers, public agencies and research institutions, model access can affect day-to-day systems, compliance plans and long-term technology strategy.
Europe’s demand, according to the source material, is not simply better service terms. Leaders want reliable access to frontier models, guarantees against sudden shutdowns, a trusted-partner scheme for non-U.S. allies, a say in where compute and chip infrastructure are located, and stronger child and youth safety rules by design.
The policy tension is that the three executives can promise cooperation but cannot remove U.S. export-control power. That makes Europe’s sovereignty agenda more concrete: the account points to a €420 billion package, gigafactories and CADA as parts of a wider push for European compute capacity and model development.
Access Control Systems: Security, Identity Management and Trust Models
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Washington Directive Reframed Dependency
The sequence began on June 12, when the Commerce Department directive cited in the source material changed Anthropic access rules. The G7 summit ran from June 15 to June 17, with the AI lunch held on the final day. That timing placed the Anthropic shutdown at the center of the discussion, even when participants spoke about safety standards, chips and democratic cooperation.
The executives presented overlapping proposals. Amodei argued for a U.S.-led coalition of democratic states, structured access for trusted partners, chip and component trade that excludes China, and joint defense against AI risks in cyber, bioterrorism and intelligence. Hassabis backed a Western coalition. Altman called for an international forum to set testing standards, saying decisions about the technology should not rest with individual labs alone.
Europe’s position, as described in the account, was more practical than philosophical: if public and private systems depend on frontier models, access needs to be durable and politically resilient.
“The democracies must not give in to the temptation to splinter.”
— Dario Amodei, Anthropic chief executive
AI model dependency management tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Access Guarantees Remain Unsettled
It is not yet clear whether the proposed trusted-partner scheme would create enforceable access rights or only a political framework. The source material says the Anthropic ban remains in place and does not describe any reversal by Washington.
Several details are still unresolved: how nationality screening would work at API scale, whether European public institutions would receive continuity protections, how child safety principles would be implemented, and how quickly European compute projects could reduce dependence on U.S.-controlled infrastructure.
MENGQI-CONTROL TCP/IP 4 Door Entry Access Control Panel Kit Electric Strike Fail Secure NO Mode Lock Enroll RFID USB Reader 110-240V Power Supply Box RFID Reader Phone APP remotely Open Door
Have smart phone APP to open lock remotely. App operate system: iPhone and Android. With Desktop USB reader,read…
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
September Talks Test Partner Model
The source material says a platform is expected within one month, with Western democratic leaders due to meet again in September. The next test is whether the Évian discussion produces concrete access rules for trusted partners, shared cyber-defense measures aimed at China-related risks, and common child safety principles.
For now, Europe’s core problem remains unresolved: the leading AI labs can negotiate, but Washington controls the policy lever that caused the shutdown.
The Developer's Playbook for Large Language Model Security: Building Secure AI Applications
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What happened at the G7 lunch in Évian?
European leaders met top AI executives, including Dario Amodei, Demis Hassabis and Sam Altman, to discuss AI access, safety and dependence after a U.S. export-control directive affected Anthropic’s model availability.
Why did Anthropic shut down model access worldwide?
According to the source material, a U.S. Commerce Department directive required Anthropic to block Fable 5 and Mythos 5 for foreign nationals. Because real-time nationality checks at API scale were not workable, Anthropic shut access worldwide.
What does Europe want from the AI companies?
European leaders want durable model access, protection against sudden shutdowns, a trusted-partner framework, more control over compute infrastructure, stronger local AI capacity and child safety rules built into deployment.
Can the AI executives give Europe those guarantees?
Only partly. The companies can offer cooperation and technical frameworks, but the source material’s central point is that U.S. government policy, not the companies alone, controls the shutdown risk.
Source: Thorsten Meyer AI
