US Directive Forces Anthropic to Shut Down Top AI Models Globally
Anthropic disabled its most advanced AI models, Fable 5 and Mythos 5, worldwide following a US government order citing national security. The directive barred foreign nationals from accessing the models, prompting Anthropic to halt all access globally due to implementation difficulties. The government's concern reportedly stems from a jailbreak incident.
Key points
- Anthropic disabled its Fable 5 and Mythos 5 AI models globally on Friday, June 12.
- The US government issued an export directive, citing national security, to bar foreign nationals from accessing these specific models.
- Anthropic stated that enforcing nationality restrictions on their shared cloud service was impractical, leading to a worldwide shutdown.
- The directive was reportedly prompted by a claimed jailbreak of Fable 5, which bypassed its safety controls.
- Anthropic considers the response disproportionate, noting similar capabilities exist in other models like OpenAI's GPT-5.5.
Artificial intelligence company Anthropic has globally deactivated its two most advanced AI models, Fable 5 and Mythos 5, following a directive from the U.S. government. The order, issued late Friday, June 12, aimed to prevent foreign nationals, both within and outside the United States, from accessing these particular models, citing national security concerns.
Anthropic announced that implementing such nationality-based restrictions on their multi-tenant cloud service would be technically challenging. Consequently, the company opted to disable Fable 5 and Mythos 5 for all users worldwide. This action marks what is believed to be the first export-control measure specifically targeting AI models themselves, rather than hardware.
The U.S. government's directive reportedly arose after a widely reported incident on June 10, where a known figure claimed to have successfully bypassed Fable 5's safety guardrails. Anthropic, however, has stated that it reviewed the alleged incident and believes the demonstrated capabilities are already present in other available models, including OpenAI's GPT-5.5. The company has described the U.S. government's response as disproportionate.
Sources
The WireByte editorial team synthesises technology news from multiple primary sources, verifies the facts, and links every source. Articles are produced with AI assistance and reviewed under our editorial policy.