Anthropic disables Fable and Mythos models after U.S. export control order
Anthropic has suspended access to its Fable and Mythos AI models following a U.S. government directive that extends export restrictions to foreign nationals in the United States, while its smaller Claude models remain available.
Anthropic announced it would disable its Fable and Mythos AI models after receiving a U.S. government directive that bars the models from use by anyone outside the United States, including foreign-national employees working in the country. The company said the order arrived at 5:21 p.m. Eastern Time and did not specify the national-security concerns prompting the action.
The company clarified that its less powerful Claude series, including the latest Claude Opus 4.8, are unaffected. "We apologize for this disruption to our customers," Anthropic posted on X, adding that it believes the directive stems from a misunderstanding and that it is working to restore access.
According to Anthropic, officials indicated the decision followed the discovery of a technique that could bypass safeguards in Fable 5, which are intended to block access to the cybersecurity capabilities of the underlying Mythos model. Anthropic contends the reported jailbreak is narrow, affecting only a single scenario, and could similarly be applied to other publicly available models such as OpenAI's GPT-5.5, which are not subject to the same export controls.
"We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company wrote. It argued that applying such a standard industry-wide would halt new model deployments and called for a transparent, fair, and technically grounded statutory process for blocking unsafe deployments.
The directive follows a series of actions by the Trump administration against Anthropic, including a February order for federal agencies to stop using its models after contract-term disputes, and a March Pentagon designation of the firm as a "supply chain risk," which the company is challenging in court. The export-control decision arrives as Anthropic prepares for a public listing, with a recent funding round valuing it at $965 billion, and could affect investor confidence.
Industry observers expressed surprise at the unprecedented move. Some commentators linked the action to broader policy tensions, while others noted that Anthropic’s own description of its Mythos model as a potential munition may have contributed to the government’s legal rationale. The episode highlights ongoing debates over how export controls and national-security considerations intersect with the rapid development of advanced AI systems.
