Anthropic faces U.S. export control dispute over Fable 5 and Mythos 5

Anthropic faces U.S. export control dispute over Fable 5 and Mythos 5
Anthropic export dispute

Anthropic is seeking talks with Trump administration officials in Washington, D.C., after a U.S. directive forces the company to suspend access to its newest artificial intelligence models for any foreign national. The order arrives days after the launch of Fable 5 and Mythos 5 and adds to a broader clash between the startup and the government over national security concerns.

Highlights

  • Anthropic suspends access to Fable 5 and Mythos 5 for all customers on Friday after receiving a U.S. government export control order citing national security authorities.
  • Defense Secretary Pete Hegseth supports blacklisting Anthropic, while ongoing litigation challenges the government's supply chain risk designation barring defense contractor use since March.
  • Anthropic claims the suspension follows concerns over a possible 'narrow, non-universal jailbreak' in Fable 5, warning the standard could halt new frontier AI deployments sector-wide.

Washington talks follow model suspension order

As first reported by CNBC, senior Anthropic staffers are meeting with Trump administration officials on Monday in an effort to resolve the latest dispute over the company’s AI products. A source close to the company says the discussions come after Anthropic receives an export control directive on Friday citing “national security authorities.”

The directive orders the company to suspend access to Fable 5 and Mythos 5 “by any foreign national, whether inside or outside the United States,” according to a company statement. Anthropic then disables access to the models for all customers to ensure compliance.

Anthropic unveils Fable 5 and Mythos 5 on Tuesday, only days before the government order. A person familiar with the discussions says the company works with government agencies to test the models before release and receives approval to deploy them.

That person says the government calls Anthropic at 1:00 p.m. ET on Friday and instructs the company to disable both models because of an unspecified national security threat. Anthropic then receives a formal letter around 5:30 p.m. ET requiring the suspension, and the same person says the company had received no earlier communication about such a threat.

National security clash deepens pressure on AI rollout

The latest action adds to a strained relationship between Anthropic and the U.S. government after an earlier dispute with the Department of Defense. The DOD labels Anthropic a supply chain risk in March, barring defense contractors from using its technology on the grounds that it purportedly threatens U.S. national security.

Defense Secretary Pete Hegseth addresses the new directive in a post on X on Saturday, saying that “every passing day” proves blacklisting Anthropic is “the right move.” Anthropic is also suing the Trump administration in an effort to reverse the supply chain risk designation, and that litigation is ongoing.

Fable 5 and Mythos 5 build on Claude Mythos Preview, a model designed to identify software security vulnerabilities. Anthropic limits the rollout under a cybersecurity initiative called Project Glasswing, and says broader availability for Fable 5 is supported by new safeguards that block responses in high-risk areas including cybersecurity and biology.

In its Friday statement, Anthropic says it believes the government’s concern centers on a “potential narrow, non-universal jailbreak,” in which a user could bypass a cybersecurity guardrail and ask Fable 5 to read a specific codebase and fix software flaws. The company says it disagrees that such a finding should trigger a recall of a commercial model and warns that applying that standard across the sector would effectively halt new deployments by frontier AI providers. Anthropic says it sees the dispute as a misunderstanding and is working to restore access as soon as possible.

In our earlier article on Amazon’s reported role in Anthropic’s AI restrictions, we covered claims that U.S. officials required Anthropic to disable access to its advanced models worldwide after security concerns were raised. That piece framed the move as a notable example of how AI safety findings can quickly translate into sweeping regulatory action affecting commercial model availability.

This material may contain third-party opinions, none of the data and information on this webpage constitutes investment advice according to our Disclaimer. While we adhere to strict Editorial Integrity, this post may contain references to products from our partners.
Weekly Top Bonuses
up to $2,500
deposit bonus for all clients
CLAIM BONUS
Your capital is at risk.