The Trump Administration Is Lifting Its Export Controls on Anthropic’s Mythos and Fable AI Models

Disclosure: Some links in this article are affiliate links. AI Maestro may earn a commission if you make a purchase, at no…

By AI Maestro July 1, 2026 2 min read
The Trump Administration Is Lifting Its Export Controls on Anthropic’s Mythos and Fable AI Models

The Trump administration is removing export restrictions on Anthropic‘s Mythos and Fable AI models after the company secured an agreement with the Commerce Department.

Commerce Secretary Howard Lutnick confirmed the change in a letter to Anthropic co-founder Tom Brown, a document seen by WIRED. The department is now clearing the way for both the Fable 5 and the more advanced Mythos 5 models. Previously, these tools were limited to specific companies and government bodies. The new rules mean no licence is needed for exporting, re-exporting, or moving these models within the country, including situations where the transfer is deemed to have happened.

The deal

These moves follow months of talks between Anthropic, the Commerce Department, and the White House. The focus was on improving safety measures to stop users from bypassing restrictions on Fable, particularly regarding cybersecurity features.

Lutnick wrote that Anthropic agreed to find and fix security risks on its own. The company also committed to working closely with the US government on protocols and release schedules for Mythos, Fable, and any future models.

Lutnick has led the administration’s push to settle the dispute with Anthropic, working alongside national cyber director Sean Cairncross.

How it happened

Anthropic initially argued the administration’s security worries were exaggerated. The company stated it was impossible to guarantee zero jailbreaks that could unlock the restricted power of the Mythos model.

In recent weeks, Anthropic changed its approach to try to bring Fable back online. This shift also changed how the company spoke to the administration. WIRED previously reported that Anthropic CEO Dario Amodei was replaced in meetings with Brown, who officials preferred on a personal level.

The company also promised to reduce the number of jailbreaks by building stronger safeguards. People familiar with the matter said this meant telling the administration what it wanted to hear rather than arguing over whether such breaches can be stopped.

What it means

Developers and businesses can now access these models without seeking special permission. This removes a significant barrier for teams looking to deploy the technology for commercial use or internal tools without navigating complex licensing paperwork.

Scroll to Top