U.S. Orders AI Blackout — Why?

Anthropic’s two most powerful AI models lasted just three days in public hands before the U.S. government ordered them shut down worldwide — and nobody outside the government fully knows why.

Story Snapshot

The U.S. government issued an export-control order on June 12, 2026, forcing Anthropic to disable its newest AI models, Claude Fable 5 and Claude Mythos 5, for all users globally.
Anthropic says it complied with the legal directive but publicly stated it disagrees that a total shutdown was necessary.
All other Claude models, including Opus 4.8, remained available — only the two newest, most capable models were pulled.
The government cited national security concerns, but the specific threat has not been made public, leaving customers and the tech world in the dark.

What Happened and When

Anthropic launched Claude Fable 5 and Claude Mythos 5 as its most capable AI models ever. The company described them as able to work on their own for longer than any previous Claude model, with major new skills in software and complex tasks. Within days of launch, the U.S. government issued an export-control directive citing national security. Anthropic pulled both models for every customer worldwide on June 12, 2026.

Anthropic posted a public notice confirming the shutdown. The company said it was “complying with the government’s legal directive” but added that it disagreed the full shutdown was warranted. Anthropic also said it was “working to restore access as soon as possible.” The company separately announced it would require 30-day data retention for all traffic on its Mythos-class models going forward.

The Government’s Case vs. Anthropic’s Response

The government’s stated concern was foreign access to the models. The export-control order specifically targeted non-U.S. citizens and organizations, according to Anthropic’s own posts. Because blocking only foreign users while keeping domestic access intact was apparently not technically feasible — or not acceptable under the directive — Anthropic cut off everyone, including paying U.S. customers.

Anthropic pushed back publicly, which is notable. The company did not simply comply in silence. Its official statement said it disagreed that the scope of the shutdown matched the actual risk. That kind of public disagreement with a government directive is unusual in the tech industry, where companies typically stay quiet when national security is invoked. It signals Anthropic believes the order went further than the evidence required.

Why This Matters Beyond AI

This story touches something bigger than one company’s product launch. The government acted on a national security claim that has not been independently verified. No outside expert has confirmed the specific threat. No court reviewed the order before it took effect. Businesses and individuals who paid for access to these tools lost them overnight, with no appeal and no timeline for getting them back. That pattern — fast government action, limited public explanation — is exactly what frustrates people across the political spectrum.

Anthropic Disables Fable 5 and Mythos 5 Worldwide After US National Security Order

The signal: Anthropic disabled Fable 5 and Mythos 5 worldwide after a US order cut foreign access on national security grounds, a live test for pre-IPO crypto bets

This is not just about hacks.…

— Fofo (@FOFOCRYPTOFR) June 13, 2026

Export controls on advanced technology are not new. The U.S. has long restricted access to powerful chips, encryption tools, and other dual-use technology on the grounds that adversaries could exploit them. Applying that same logic to AI models is a natural next step as these tools grow more powerful. The harder question is whether the government’s process is transparent enough. When a product is shut down globally in three days based on a classified or unexplained threat, the public has no way to judge whether the response was proportionate. That lack of accountability is a concern worth watching, no matter which party is in power.