Anthropic has cut off access to its two most advanced AI models, Fable 5 and Mythos 5, for all users worldwide following a U.S. government export control directive that mandates the company block access by any foreign national.
US Gov: bar 'foreign national' access to Fable 5
The directive, which Anthropic says it received at 5:21pm ET on June 12, invokes "national security" authorities and prohibits access to both models by foreign nationals within or outside the United States, including Anthropic's own foreign-national employees.
The company says the net effect of the order is that it must disable both models for all customers in order to comply. All other Anthropic models, including Claude Opus 4.8, remain unaffected.
The timing is inconvenient. Anthropic began rolling out Fable 5 on June 9, offered at no cost to all Pro, Max, and Enterprise customers through June 22. The model that was given away to millions for free three days ago is now unavailable for everyone.
Fable 5 is the safeguarded counterpart of Mythos 5. Both share the same underlying model, but Fable incorporates additional safeguards.
Fable blocks or redirects sensitive cybersecurity, biology, and chemistry inquiries, while the unrestricted Mythos 5 is available only to vetted government cyberdefenders and life sciences partners.
In a developer notice, Anthropic stated that new sessions would revert to a user's default model or Opus 4.8, existing Fable 5 sessions would terminate with an error, and Platform requests to Fable 5 would also fail. It advised integrators to transition to other models.
UK's Minister for AI and Online Safety, Kanishka Narayan MP, noted that the pause affected customers in both the US and UK, framing it as an argument for technological sovereignty and referencing the government's £1.1bn AI chip investment.
Anthropic's interpretation is that the order stems from a reported method to jailbreak Fable 5. It says it examined a demo and found only minor, already-known bugs, the kind that other publicly-available models can discover without any bypass.
"To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws," states Anthropic.
"Our understanding is that one potential jailbreak was shared with the government."
"We are complying with the government's legal directive and are removing access to Fable 5 and Mythos 5 for all users. 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."
The company asserts that the capability is widely accessible elsewhere, pointing to OpenAI's GPT-5.5, and is used by defenders on a daily basis.
Anthropic maintains that the order is based on a misunderstanding and says it is working to restore access, while promising more details within 24 hours.
Test every layer before attackers do
Security teams log 54% of successful attacks and alert on just 14%. The rest move through your environment unseen.
The Picus whitepaper shows how breach and attack simulation tests your SIEM and EDR rules so threats stop slipping by detection.
Related Articles:
Anthropic rolls out Claude Fable 5, but it's available for a limited time
Anthropic confirms Claude Mythos-class models will roll out to the public
Anthropic’s restricted Claude Mythos model may be coming to Claude Code
XBOW tests Anthropic's Mythos Preview for offensive security
Claude AI skills are becoming valuable fast—this learning course is just $20 right now