After weeks of intense negotiations, the White House has allowed Anthropic to restore access to its most advanced AI model, Mythos, for a select group of US companies and government agencies. The decision, communicated in a letter from Commerce Secretary Howard Lutnick, marks a partial easing of restrictions imposed by the Trump administration, but leaves the fate of the consumer version Fable 5 uncertain.
The letter, addressed to Anthropic co-founder and chief compute officer Tom Brown, states that access to Mythos is permitted only for trusted partners, after adequate safeguards have been implemented. Lutnick wrote: Anthropic has worked with the U.S. government to address risks associated with the Covered Models. These efforts have yielded significant progress. However, the government stopped short of authorizing a broader rollout and said nothing about Claude Fable 5, the consumer-facing version of Mythos, which Anthropic had released with additional protections.
Sponsored Protocol
The Dispute with the White House and Security Concerns
This partial reinstatement comes roughly two weeks after the Trump administration sent an export control directive to Anthropic, requiring the company to limit foreign nationals' access to Mythos and Fable 5, including those living and working in the United States. In response, Anthropic disabled access to the models entirely. Now, with the new letter, Lutnick has authorized approved organizations to allow their foreign national employees to use Mythos, and Anthropic may do the same for its own foreign national staff.
The tension escalated after the administration learned that Anthropic had granted access to a South Korean telecommunications firm believed to have ties to China. Additionally, Amazon and the National Security Agency separately raised concerns that Fable 5 could be jailbroken. These events convinced officials to take action. To resolve the situation, Anthropic sent senior members from its cybersecurity and AI safety teams to Washington, including Tom Brown and public policy chief Sarah Heck.
Sponsored Protocol
Restoring Mythos 5 is a promising step, but the saga raises broader questions about the direction of US AI policy, with increasing government control. On the same day, OpenAI announced it was delaying the release of its GPT-5.6 models at the request of the Trump administration. Dean Ball, head of the strategic futures team at OpenAI and a former White House AI adviser, noted that the dispute showed frontier AI developers need an explicit green light from the government. The situation has been costly for Anthropic, which has already sued the administration over a supply chain security dispute.
Sponsored Protocol
For comparison, OpenAI tightens access in Europe — Italian SMEs left holding the bag? shows how other companies face similar restrictions. Also, Paul Meade Leaves Apple for OpenAI indicates a consolidation of expertise in the sector. The Anthropic-White House battle could set a precedent for future AI model releases. According to WIRED, discussions will continue over the weekend, with both sides hopeful for a lasting policy framework.
Source: https://www.wired.com/story/anthropic-restores-access-to-mythos
