Commerce dept. fears a Fable 5 "jailbreak" could pose a national security risk.
Anthropic completely cut off access to its Mythos 5 and Fable 5 models on Friday evening, just days after their launch.
This action follows Anthropic's receipt of a directive from the US Commerce Department on Friday evening, which placed the new models under export controls limiting their use exclusively within the United States. In a message posted Friday night, Anthropic stated that the only way to immediately comply with that government order "is that we must abruptly disable Fable 5 and Mythos 5 for all our customers." Access to other Anthropic models remains unaffected.
An Axios report cited an administration official stating that the administration is troubled by reports of a jailbreak that allegedly bypasses broad classifier-based safeguards designed to block Fable 5 prompts related to cybersecurity, chemistry, and biology. The administration reportedly requested a pause in the release of these models to allow time for the "national security apparatus" to be "hardened" against such a threat. That hardening could be completed "in the next few weeks," according to Axios' source.
In its Friday night announcement, Anthropic said the government has only provided it with "verbal evidence of a potential narrow, non-universal jailbreak" that involves getting Fable 5 to review a specific codebase for software flaws. The company says it has only seen evidence of this kind of jailbreak being used to find "minor" and "relatively simple" software vulnerabilities, and that other publicly available models like GPT-5.5 have similar capabilities in this regard.
"We are complying with the government's legal directive and are removing access to Fable 5 and Mythos 5 for all users," Anthropic writes. "However, 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. If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers."
Earlier this month, President Trump signed an executive order urging AI model makers to submit to voluntary government security testing. That order came after an initial signing ceremony planned for last month was abruptly postponed amid reported concerns of internal disagreements about it within the administration.
Anthropic apologized to customers for a "disruption" that it said results from a "misunderstanding," and stated that it will release more details about the situation in the next 24 hours.