Trump says Anthropic talks ‘going fine’ at G7 AI lunch

Trump told reporters at the G7 summit in Évian-les-Bains that negotiations with Anthropic over its export-controlled AI models are “going fine,” per Reuters.marketscreener
The Commerce Department banned foreign access to Anthropic’s Claude Fable 5 and Mythos 5 models on June 12, citing national security concerns after Amazon CEO Andy Jassy flagged potential cybersecurity risks.techcrunch
Wednesday’s G7 working lunch on AI — the first to include leaders of Anthropic, OpenAI, and Google DeepMind — comes as European allies push for a “trusted partners” framework to reduce dependence on U.S. AI providers.apnews

Trump Says Anthropic Talks ‘Going Fine’ as AI Model Ban Drags On

President Trump told reporters at the G7 summit in Evian-les-Bains, France, that negotiations with Anthropic are “going fine,” offering his first public remarks on the standoff since his administration ordered the AI company to disable its most advanced models nearly a week ago. The comment came as Trump appeared alongside Anthropic CEO Dario Amodei and other AI executives at a Wednesday working lunch on artificial intelligence — an encounter freighted with tension given the unresolved export controls hanging over the company.youtube

The Shutdown and Its Origins

The crisis began on Friday, June 12, when Commerce Secretary Howard Lutnick sent Anthropic a letter invoking the 2018 Export Control Reform Act to bar all foreign nationals, inside or outside the United States, from accessing the company’s Claude Fable 5 and Mythos 5 models. The order marked the first time the Commerce Department had used that authority against a commercially available AI product. Because Anthropic could not readily distinguish foreign users from American ones, the company said the “net effect” was to “abruptly disable” both models for all customers worldwide.businessinsider

The trigger, according to multiple reports, was Amazon CEO Andy Jassy, who privately warned Treasury Secretary Scott Bessent and other officials that Amazon researchers had used Fable 5 to extract information that could aid cyberattacks. Anthropic has disputed the severity of the vulnerability, calling it a “potential narrow, non-universal jailbreak” and noting that similar capabilities exist in other publicly available models.techcrunch

Washington Deadlock, Congressional Doubt

Anthropic dispatched technical staff to Washington over the weekend and met with administration officials on Monday, but the talks ended without the export controls being lifted. An Anthropic spokesperson said “both parties are working quickly to get this resolved”.instagram

On Capitol Hill, lawmakers from both parties questioned the order’s rationale. Sen. Angus King said he was “a little skeptical because of their otherwise announced antipathy to this company,” referencing the broader feud over Anthropic’s refusal to allow its AI to be used for domestic surveillance or autonomous weapons. Rep. Bennie Thompson called the approach “ad hoc,” while House Homeland Security Chairman Andrew Garbarino cautioned against responses that “unintentionally disadvantage American companies” or allied partners.cyberscoop

Europe Pushes for AI Sovereignty

The ban has amplified European calls for technological independence. Fortune reported that the shutdown confirmed what some officials had called a U.S. “kill switch” over foreign access to American AI systems. At the G7, European allies proposed a “trusted partners” framework that would grant allied nations access to advanced U.S. AI models, according to Reuters, with French President Emmanuel Macron leading the push. Wednesday’s AI-focused session brought together the CEOs of Anthropic, OpenAI, Google DeepMind, and Mistral AI alongside world leaders — the first G7 gathering to include all three leading American AI firms.fortune

With no resolution in sight, the episode has underscored the fragility of global dependence on a handful of U.S. AI providers — and the political risks that come with it.