Anthropic disables Fable 5 and Mythos 5 after US export control directive

Anthropic said on 12 June 2026 that it is disabling access to its two flagship models, Fable 5 and Mythos 5, for all customers after receiving a US government export control directive. The order suspends access to both models by any foreign national, whether inside or outside the United States, including foreign national employees of Anthropic itself.

What the directive requires

According to the company, the directive arrived at 5:21pm ET and cites national security authorities, without spelling out the specific concern behind it. Because Anthropic cannot practically separate foreign nationals from the rest of its customer base, it says it has to abruptly disable Fable 5 and Mythos 5 for everyone in order to comply. Access to all other Anthropic models is unaffected.

Anthropic says the government believes it has learned of a method of bypassing, or "jailbreaking", Fable 5. The company reviewed a demonstration of the technique and says it was used to identify a small number of previously known, minor vulnerabilities. Those flaws appear relatively simple, and other publicly available models can discover them as well without requiring any bypass.

Safeguards and testing

The company pointed to pre-launch red-teaming carried out with the US government, the UK AI Safety Institute, external organisations and internal teams, amounting to thousands of hours in total. Those tests, it says, found no universal jailbreak and showed Fable 5's safeguards to be substantially more effective than those of any previously deployed model. Anthropic also cited its defense-in-depth approach and its 30-day retention of customer data, a policy it says allows it to research and mitigate jailbreak attempts.

Disagreement over the standard

Anthropic says the government has so far provided only verbal evidence of a narrow, non-universal jailbreak, which it describes as essentially asking the model to read a specific codebase and fix software flaws. The company says it validated that the capability demonstrated there is widely available from other models, including OpenAI's GPT-5.5, and is used every day by the defenders who keep systems safe.

Anthropic said it is complying with the legal directive but disagrees that such a finding should be cause for recalling a commercial model deployed to hundreds of millions of people, warning that applying that standard across the industry would essentially halt new model deployments at every frontier provider. It repeated its position that governments should be able to block unsafe deployments as part of a statutory process that is transparent, fair and grounded in technical facts, and apologised to customers, saying it is working to restore access as soon as possible.