Washington just proved it can turn off a top AI model overnight, and that should worry anyone who values free markets and limited government.

Quick Take

• Anthropic said the U.S. government ordered it to suspend access to Fable 5 and Mythos 5 for foreign nationals.^[2]
• The company said it had to disable both models for all customers to stay in compliance.^[2][3]
• Anthropic also said access to its other Claude models was not affected.^[2]
• The company said the government’s concern involved a possible jailbreak, but it called the move a misunderstanding.^[1][2]

What Happened

Anthropic said the U.S. government issued an export control directive that forced it to suspend access to Fable 5 and Mythos 5.^[2] The company said the order covered foreign nationals inside and outside the United States, including foreign national employees. Anthropic said that rule made a selective rollout too hard to manage, so it disabled both models for everyone.^[2][3]

The company said it received the directive at 5:21 p.m. and moved fast to comply.^[1][3] Anthropic also said the letter did not explain the national security concern in detail.^[1][2] That matters because the public record shows a government action first, then a broad shutdown. The company did not present this as a routine product pause or a normal safety update.^[2][3]

Why Conservatives See a Bigger Problem

This story fits a broader pattern of growing federal control over advanced technology. Recent policy material shows the United States has used export controls more aggressively on AI chips and related tools since 2022. That matters because export controls were once aimed at physical goods, not cloud-based software access. When the government starts policing who may touch an AI model, the line between national security and overreach gets thin fast.

Anthropic’s own statement also leaves room for debate about the real cause. The company said the government may have acted because of a possible jailbreak, and it said it reviewed a demonstration of that technique.^[1][2] Anthropic then said the reported vulnerabilities were minor and already discoverable by public models.^[1] That is not the same thing as proving the shutdown was only a safety decision. The public evidence still points most strongly to a government directive.^[2][3]

The Safety Argument Is Not Nothing

There is a real safety case behind some of these restrictions. Public research and security guidance say frontier models can create new risks, resist shutdown, and support harmful misuse if access is not controlled. Those warnings help explain why officials may have acted quickly. But broad risk talk is not the same as a clear technical record tied to this exact shutdown. The materials available here do not show a public incident report, red-team file, or internal review that proves a purely voluntary internal decision.

On Friday night, the tech world saw something completely unprecedented: the global, sudden shutdown of Anthropic’s Claude Fable 5 and Mythos 5.

Not throttled. Not fenced. Switched off globally by an undisclosed government export-control directive—just three days after launch.…

— ARAGS Inc. (@ARAGS_Inc) June 15, 2026

The strongest fact remains simple: Anthropic said the government ordered the suspension, and the company complied.^[2] The company also said the move was a misunderstanding and that it was trying to restore access.^[2][1] Until more documents surface, the public record supports a mixed picture, but not the claim that Anthropic acted only on its own internal judgment.^[1][2][3]

Sources:

[1] Web – Why Did the Smartest AI in the World Just Go Dark?

[2] Web – Statement on the US government directive to suspend access to …

[3] X – Anthropic