# Washington Orders Anthropic to Cut Off Foreign Access to Its Top AI Models

An export-control directive forced the company to disable its Fable 5 and Mythos 5 systems for all foreign nationals, the first such order aimed at specific artificial-intelligence models.

- Published: 2026-06-13T11:12:07.337Z
- Canonical: https://polylog.news/2026-06-13/washington-orders-anthropic-to-cut-off-foreign-access-to-its
- Publisher: Polylog (Global desk)
- Section: tech
- Sources: [Al Jazeera](https://www.aljazeera.com/news/2026/6/13/us-orders-anthropic-to-disable-ai-models-for-all-foreign-nationals), [Deutsche Welle](https://www.dw.com/en/anthropic-cuts-top-tier-ai-access-after-us-foreigner-ban/a-77534688), [South China Morning Post](https://www.scmp.com/tech/article/3356990/us-blocks-foreign-access-anthropics-newest-ai-models-over-security-risks)

The United States government has ordered the artificial-intelligence company Anthropic to block all foreign nationals from using its most advanced models, Fable 5 and Mythos 5, on national-security grounds. The company [said](https://www.aljazeera.com/news/2026/6/13/us-orders-anthropic-to-disable-ai-models-for-all-foreign-nationals) it received an export-control directive and responded by disabling customer access to both models entirely while it sought to comply.

The order applies to any foreign national inside or outside the country, including Anthropic's own foreign employees, according to [Deutsche Welle](https://www.dw.com/en/anthropic-cuts-top-tier-ai-access-after-us-foreigner-ban/a-77534688). The government's stated concern, conveyed largely as verbal evidence, is that a safeguard meant to stop Fable 5 from helping identify software vulnerabilities can be bypassed, and that the same weakness likely exists in rival systems.

Anthropic [disputed](https://www.scmp.com/tech/article/3356990/us-blocks-foreign-access-anthropics-newest-ai-models-over-security-risks) the action, warning that recalling a model used by hundreds of millions of people over a narrow potential jailbreak, if applied across the industry, would halt new model deployments by every frontier developer. The measure is the first United States export control directed at named artificial-intelligence models rather than chips or code.

## What this means

This extends the logic of technology decoupling from hardware to the models themselves, treating frontier artificial intelligence as a controlled strategic good. For a global company, it raises the cost of serving an international customer base and hands an opening to non-United States developers. It also signals that model access can now be switched off by directive, a precedent that reshapes how foreign governments and firms plan their own systems.

## What to watch

- Whether the directive is broadened to other United States model developers
- Responses from Chinese and European firms positioning as alternatives
- Any legal challenge or carve-out for allied governments and researchers