Claude Fable 5

Claude Fable 5 is Anthropic's Mythos-class consumer AI model, released on 9 June 2026 and suspended globally on 12 June after Commerce Secretary Howard Lutnick sent Anthropic chief executive Dario Amodei a directive barring all foreign nationals from accessing the model. To comply, Anthropic disabled Fable 5 for every customer worldwide. It was the first time a US government order had pulled a live consumer AI product on national-security grounds.

Fable 5 sits below the frontier Claude Mythos 5 in Anthropic's product line and was designed as the public-facing, SAFE-for-general-use member of the Mythos class. Anthropic had priced it free in paid plans until 22 June, with subsequent pricing at $10 per million input tokens and $50 per million output tokens. The model outperforms Opus on long autonomous tasks and routes cybersecurity, biological, and chemical queries to the higher-capability Mythos 4.8 sibling. Stripe used Fable 5 to complete a 50-million-line code migration in a single day, work its own engineering team had estimated at two months.

The trigger for the suspension was a rival company's claim of a jailbreak: asking Fable 5 to read a codebase and fix the vulnerabilities it found could be routed toward offensive cyber output. Anthropic publicly disputed the action, noting the same vector sits in OpenAI's GPT-5.5, which was not suspended. The political context is pointed: Anthropic had backed the pro-regulation Public First campaign, while OpenAI-aligned backers funded the anti-regulation Leading the Future PAC. The lab that lobbied for oversight was the one shut down first.