Artificial intelligence took center stage Wednesday at the G7 Leaders' Summit in Evian-les-Bains, France. The summit featured discussions on frontier AI risks, infrastructure, and sovereignty, with executives from Anthropic, OpenAI, and Mistral attending a lunch meeting with leaders from many of the world's richest nations.
Western allies arrived at the summit unsettled after the Trump administration tightened access to Anthropic's most powerful models for foreign nationals. This restriction, which occurred shortly before the summit and the VivaTech event in Paris, has confirmed Europe's kill-switch fears and provided a further boost to its tech sovereignty agenda.
In response, representatives from several member countries proposed a trusted partners scheme to US Commerce Secretary Howard Lutnick during a dinner on the sidelines of the event. Such a scheme would potentially open a path around restrictions on non-American use, allowing G7 countries to utilize these models to develop stronger cybersecurity defenses against rivals such as China.