Anthropic shuts down its most powerful new AI models, Claude Fable 5 and Mythos 5, following a US government export control directive.
Just 72 hours after launching its new frontier "Mythos-class" AI models—Claude Fable 5 and Claude Mythos 5—Anthropic has taken them offline globally. The sudden shutdown was triggered by a US government export control directive issued on June 12, 2026, which restricted model access by foreign nationals inside or outside the United States. Since Anthropic could not filter users by nationality in real-time, it disabled the models completely. While the government's order is linked to a narrow security "jailbreak" method that bypasses Fable 5's safety filters, Anthropic has publicly contested the decision, warning that such overly broad actions risk stalling all frontier model deployments.
The US government's immediate shutdown of Claude Fable 5 proves that national security directives, rather than market forces or internal alignment, are now the ultimate governors of frontier AI.
- –The U.S. government's action shows it is willing to prioritize national security over market availability, establishing a heavy-handed precedent for future frontier model deployments.
- –Nationality-based export controls on software are functionally unenforceable for cloud-based AI APIs, forcing companies into all-or-nothing global shutdowns.
- –Anthropic’s public disagreement with the decision highlights a growing rift between AI labs seeking rapid deployment and regulatory bodies acting on narrow vulnerabilities.
DISCOVERED
1h ago
2026-06-14
PUBLISHED
2h ago
2026-06-14
RELEVANCE
AUTHOR
kekkodamato_
