Anthropic announces the US Commerce Department’s recent decision to lift export controls on its Claude Fable 5 and Mythos 5 AI models, just under three weeks after the firm was mandated to halt access to its cutting-edge models due to national security concerns.
“We will commence the restoration of access tomorrow,” Anthropic conveyed via a statement on X.
The company had previously deactivated its Mythos 5 and Fable 5 models promptly following the export-control directive issued on June 12.
We have received notice that the Department of Commerce has rescinded export controls on Claude Fable 5 and Mythos 5.
We’ll begin restoring access tomorrow and will provide further updates soon.
We appreciate our users’ patience and everyone who assisted us on…— Anthropic (@AnthropicAI) June 30, 2026
On Friday, Anthropic indicated that the US government had permitted the release of its Claude Mythos 5 AI model to select “trusted” US organizations, partially reversing the previous order.
“In the past fortnight, we have collaborated closely with Anthropic to evaluate and sanction Fable 5 to ensure coherence across the US Government and bolster America’s preeminence in AI,” stated US Commerce Secretary Howard Lutnick in a message on X.
Washington has intensified scrutiny over new model launches to pinpoint potential hazards posed by advanced AI systems that are fueling the AI renaissance and substantial capital infusion.
Over the past fortnight, we have worked intimately with Anthropic to assess and validate Fable 5 to ensure alignment across the US Government and enhance America’s leadership in AI. https://t.co/PtVJWyQ9kH
— Howard Lutnick (@howardlutnick) June 30, 2026
However, the government’s vetting process for companies seeking access to these models has sparked criticism.
OpenAI CEO Sam Altman remarked last week on X that while extensive safety evaluations “are not a negative concept,” he is resistant to the notion of government dictating customer access.

ChatGPT developer OpenAI postponed the comprehensive public debut of GPT-5.6 at the behest of the US government, restricting access to a select cohort of vetted partners.
Source link: Startupdaily.net.






