Trump’s Anthropic shutdown just made the case for non-American AI

The U.S. government demanded Anthropic disable access to its latest AI models, Fable 5 and Mythos 5, for all foreign users, including its own employees. The company complied over the weekend, citing White House instructions, and the shutdown reignited debates about global dependence on American AI infrastructure. The abrupt restriction on two of the most advanced models—already limited in high-risk applications—has accelerated discussions about national AI sovereignty. Critics argue the U.S. wields disproportionate control over critical technology, raising concerns about geopolitical leverage and security risks. In the UK, AI and online safety minister Kanishka Narayan used the incident to emphasize the need for domestic AI capacity, framing it as a matter of national sovereignty. He warned that Britain must decide how AI shapes its future before external powers dictate the terms, comparing the threat to traditional security challenges. France’s response was more direct, with presidential candidate Gabriel Attal calling the shutdown the start of an ‘AI war.’ He compared the move to a strategic blockade, urging France to reduce reliance on foreign AI providers. The French political spectrum broadly shares this alarm, with media outlets like *Le Monde* highlighting the urgency of European tech independence. The controversy underscores broader unease in Europe about U.S. reliability, particularly under the Trump administration. Trade disputes and NATO tensions have further fueled concerns, with Attal positioning AI as a central issue in France’s upcoming presidential elections. The incident has also intensified EU efforts to reduce dependence on external tech providers, including in AI, cloud computing, and semiconductor manufacturing. While Anthropic may restore access to its models, the shutdown has exposed vulnerabilities in global AI governance.