US assures allies on Anthropic AI access at Pax Silica Summit

The European Union and India have both opened diplomatic channels with the Trump administration over access to Anthropic's most advanced AI models, after a U.S. export control directive forced the company to disable Fable 5 and Mythos 5 for all users worldwide earlier this month.

Brussels Raises Concerns With Washington

European Commission Executive Vice President Henna Virkkunen raised the issue directly with the White House during a trip to Washington this week, according to multiple reports. The abrupt cutoff left European developers and partner organizations without access to the models they had begun integrating just days after their launch on June 9.aiweekly+1

A U.S. Commerce Department letter, signed by Secretary Howard Lutnick, ordered Anthropic on June 12 to suspend access to Fable 5 and Mythos 5 for any foreign national regardless of location, including the company's own non-citizen employees. Anthropic confirmed it took both models fully offline because real-time nationality verification was not feasible.forbes+1

Commission spokesperson Thomas Regnier said the EU was examining the directive's implications and pointed to the AI Office's enforcement powers, set to take effect in August 2026, as a potential avenue to "ensure to receive, if needed, model access".aiweekly

India Seeks Assurances at Pax Silica Summit

Separately, U.S. Under Secretary of Economic Affairs Jacob Helberg confirmed on Thursday at the second Pax Silica Summit that the United States and India are engaged in what he called "very sensitive national security discussions" over a gradual rollout of Anthropic's Fable model to trusted partners.timesofoman+1

"We intend to continue a gradual measured approach to how we release Anthropic's models in a way that is safe, both for ourselves but also for our Indian counterparts as well as all our trusted partners," Helberg told news agency ANI. India was among 35 nations that signed a statement on AI opportunity at the summit.theprint+1

Indian officials said the U.S. provided assurances that "for trusted partners access will not be an issue" going forward, and that future models would not be cut off abruptly.theprint

National Security Rationale Remains Disputed

The Trump administration has not publicly detailed the national security rationale behind the order, though reports indicate officials became aware of a method to "jailbreak" Fable 5's safety filters — specifically circumventing protections on advanced cybersecurity capabilities built into the underlying Mythos framework. Anthropic has called the restriction a misunderstanding and said it plans to share more details soon. Other Anthropic models, including older versions, remain accessible.aljazeera+2