Anthropic has suspended access to its Fable 5 and Mythos 5 AI models after the U.S. government imposed export restrictions on June 12. The Commerce Department barred foreign nationals and users outside the U.S. from accessing the models, citing national security concerns. The move follows reports that Amazon researchers discovered security vulnerabilities through jailbreak testing, prompting the Trump administration to intervene.
Part 1: Immediate Action & Core Facts
- Anthropic disabled access to Fable 5 and Mythos 5 on June 12 after the U.S. government imposed restrictions.
- The Commerce Department barred foreign nationals and non-U.S. users from accessing the models, citing national security risks.
Part 2: Deeper Dive & Context
Government Rationale
The Commerce Department restricted access due to concerns over potential misuse of the AI models. Commerce Secretary Howard Lutnick notified Anthropic CEO Dario Amodei of the decision, which prohibits foreign nationals and users outside the U.S. from accessing the models.
Amazon’s Role
Reports suggest Amazon researchers conducted jailbreak testing on Anthropic’s models, revealing security vulnerabilities. Katie Moussouris, CEO of Luta Security, argued the findings were more useful for cybersecurity defense than exploitation. Amazon acknowledged consulting with the government on security risks but did not confirm its involvement in the testing.
Industry Impact
The restrictions raise questions about how AI advancements may trigger regulatory scrutiny. Anthropic had previously advocated for stronger AI oversight, which some analysts see as ironic given the government’s decision. The move could affect future AI investments and IPO valuations for companies like OpenAI and Anthropic.
Competing Perspectives
Some experts argue the restrictions are an overreaction, while others see them as necessary to prevent misuse. The White House has not publicly detailed its evaluation process, leaving the rationale for the ban open to interpretation.