US lifts export restrictions on Anthropic’s Mythos and Fable AI models

The United States has dropped a requirement that forced Anthropic to obtain a license before exporting its Mythos and Fable models abroad, effectively cutting off public access to what are widely considered the most advanced AI models to date. The AI lab announced it would begin restoring access to the models on Wednesday, July 1.

On June 12, the US government had added the products to its list of export-restricted technologies, meaning they could no longer be made available to foreign nationals without special approval. Complying with that rule proved impractical at scale, forcing Anthropic to end public access altogether.

After weeks of talks, Secretary of Commerce Howard Lutnick said Anthropic “has agreed to proactively detect and address security risks associated with the models; to work diligently with the U.S. government on protocols and standards and releases for Mythos, Fable and future models; and to inform the US government of any malicious activity.” Anthropic had already publicly pledged to do much of this voluntarily months before the export rule existed, partly explaining why cybersecurity experts were skeptical of the restrictions. To them, the ban looked less like a security fix and more like leverage—a way for the Trump administration to punish Anthropic for its executives’ public criticism of how the government and the president’s political opponents might use the technology.

Mythos was originally made available to a select group of organizations in April to allay concerns about its ability to identify and exploit software vulnerabilities, while Fable was released to the public in June with additional security guardrails. However, with Asian AI companies releasing their own models approaching Mythos-level capabilities—including Fugu and Tulonfeng—the US government faced pressure to ease restrictions on Anthropic to ensure American AI could compete globally. Last week, Lutnick cleared Mythos for release to select customers approved by the White House. OpenAI’s latest models were also released to a group of organizations approved by the Trump team rather than the public.

The Trump administration’s erratic approach to AI policymaking has left companies across the industry uncertain about future model release rules. An executive order issued in June, signaling a desire to review models before release, drew criticism from influential analysts like Dean W. Ball, who recently started a policy position at OpenAI.