The Trump administration on Tuesday reversed its earlier decision to restrict two of Anthropic's newest artificial intelligence models, lifting export controls on Fable 5 and Mythos 5 just over two weeks after they were abruptly pulled from public access.
Commerce Secretary Howard Lutnick announced the move on X, saying his agency had “worked closely with Anthropic to analyze and approve Fable 5 to ensure alignment across the US Government and strengthen America's leadership in AI.” Anthropic confirmed it received official notification and will begin restoring the models Wednesday.
“We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models,” the company wrote on X.
The about-face comes after the Trump administration issued a federal export control order earlier this month, forcing Anthropic to yank its Claude Mythos 5 and Fable models just three days after their debut. According to WIRED, Lutnick informed Anthropic co-founder Tom Brown of the reversal in a letter Tuesday, stating that “a license is no longer required for the export, reexport, or in-country transfer, including deemed export or deemed reexport, of the Mythos or Fable models.”
The original restrictions were triggered after Amazon CEO Andy Jassy raised concerns with the administration about a potential method to “jailbreak” the guardrails on Fable, according to multiple outlets. Amazon, a major investor in Anthropic, flagged the vulnerability before the Commerce Department moved to impose controls. A source close to Anthropic told The World Signal that the company was given just 90 minutes to comply with the order, with no prior warning of any national security threat.
The controversy centers on the two models' differing safety profiles. Fable 5 was released to the public with safeguards designed to prevent misuse, while Mythos 5—based on Anthropic’s first Mythos model, which the company had initially withheld over fears it could supercharge hacking capabilities—was provided to a small group of cyber-defenders and infrastructure providers with fewer guardrails. Tuesday's decision follows the government's approval of Claude Mythos 5 for a limited set of companies just days earlier.
The episode highlights the Trump administration's evolving approach to AI regulation, balancing national security concerns with industry pressure. It also underscores the influence of major tech investors like Amazon in shaping policy. The reversal comes amid broader political battles, including the Supreme Court's recent ruling on birthright citizenship and ongoing disputes over Pentagon funding, as the administration navigates a complex regulatory landscape.
Anthropic has not commented on whether the restored models will include any new safeguards or modifications following the review. The company's decision to resume access Wednesday suggests the Commerce Department's analysis found no unresolved security risks.
