US lifts block on Anthropic's Claude Mythos 5 for more than 100 vetted US institutions

The US government on Friday lifted its block on Anthropic's powerful Claude Mythos 5 model, clearing the company to release it to more than 100 US institutions, including major corporations and government agencies, Semafor reported. The decision was set out in a letter sent to the AI lab on Friday afternoon.

The move marks a major de-escalation in a confrontation between the Trump administration and one of the world's most valuable private companies, which began two weeks ago when Washington imposed export controls on the model.

What the letter says

Commerce Secretary Howard Lutnick wrote to Anthropic's chief compute officer Tom Brown that "appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model," citing "significant progress" in the intense, daily talks between the government and the company since the block took effect. Anthropic has committed to work with the US government on protocols and standards for its model releases, Lutnick added.

Under the arrangement, a license is no longer required to export, reexport or transfer the model in-country to the entities listed in an annex attached to the letter and their foreign national employees — or to Anthropic's own foreign national employees.

The letter is silent on Fable 5, a weaker sibling model that was briefly the most powerful AI system widely available to consumers. People close to the talks said the two sides are moving toward releasing Fable as well, though the timeline is unclear.

A new regulatory regime

The block had shut down both Mythos and Fable after Amazon and other companies warned the models could be jailbroken for malicious purposes. Lutnick's letter now marks the beginnings of a framework that gives the US government a direct say over when frontier AI models reach users.

Commerce Department spokesman Benno Kass pointed to how quickly the dispute was resolved: "In just two weeks, we have worked diligently to ensure America remains the global leader in AI while safeguarding our security," he said.

The same day, Anthropic's leading competitor OpenAI released its latest model, GPT-5.6, to a short list of government-approved partners.

Allies left waiting

The oversight framework is being built on the fly, and many users of the most capable tools — from foreign governments and companies to ordinary consumers — still do not know when they will gain access to Mythos or Fable. European officials and other US allies have expressed frustration at their new dependence on decisions made in Washington.

As Semafor previously reported, the initial restrictions followed concerns that Mythos had been released to partners too closely linked to China, reportedly a South Korean telecommunications provider.