On June 18 local time at the G7 summit held in Evian, France, U.S. President Trump stated that negotiations between the U.S. government and Anthropic regarding the restoration of access to its latest artificial intelligence models were “proceeding smoothly.” This statement came as leaders from multiple countries and executives from tech companies discussed access to advanced AI models, technological safety, and global governance issues. Following a working lunch with G7 leaders and AI industry executives, Trump told reporters that the meeting yielded positive results and emphasized that artificial intelligence was driving substantial investment flows into the United States. “We had deep discussions with some of the most important people in the AI field; it was a very successful meeting,” Trump said, adding that the United States was building the world’s largest AI-related infrastructure. The controversy stemmed from the U.S. government’s imposition of overseas access restrictions on two of Anthropic’s advanced models last week. In response, Anthropic suspended service related to these models for all users, sparking widespread international concern over equitable access to advanced AI resources and global coordination mechanisms. During the summit, AI industry leaders including OpenAI CEO Sam Altman, Google DeepMind head Demis Hassabis, and Anthropic CEO Dario Amodei attended the meetings. French President Emmanuel Macron publicly expressed concerns about unilateral restriction measures by the United States. He noted that while it is understandable for the U.S. to remain vigilant against AI risks, if access to critical technologies can be cut off at any moment, other nations and enterprises would find it difficult to rely confidently on such services. “If you could lose access overnight, we wouldn’t choose those companies’ models,” Macron said. The U.S. government has not yet disclosed specific security considerations behind the restrictions. However, many industry analysts believe that certain capabilities present in the restricted models also exist in other publicly available large language models, suggesting that comprehensive restrictions on overseas access may constitute an overreaction. According to sources familiar with the matter, several industry leaders urged G7 nations to strengthen cooperation on AI safety governance during the conference. Amodei and Hassabis advocated for leadership by the United States in developing AI safety standards, while some attendees suggested referencing the Financial Stability Board established after the 2008 financial crisis to create a more multilateral international coordination mechanism. Altman called for the establishment of an international forum to promote globally recognized AI testing standards and independent risk assessment systems. He stressed that while addressing potential AI risks, care should be taken to avoid excessive concentration of advanced technologies and decision-making power within a few institutions. Aidan Gomez, CEO of Canadian AI company Cohere, stated there was general consensus among participants that G7 nations needed tighter coordination mechanisms to jointly address the opportunities and challenges brought by artificial intelligence. Later that day, Trump remarked at a press briefing that artificial intelligence “could bring immense value but also poses risks,” noting that countries must exercise caution while promoting technological development.