Anthropic shuts down Mythos access after sweeping U.S. order

Anthropic has shut down access to its most advanced AI models, Mythos and Fable 5, after receiving an order from the U.S. government. The directive, issued by the Trump administration, requires the company to prevent any foreign national—whether inside or outside the U.S.—from accessing these systems. A U.S. official confirmed the Commerce Department sent the letter, and Anthropic complied by cutting off access to all customers to ensure compliance with the order. This marks the first time the U.S. has imposed such sweeping restrictions on AI software developed by an American company. Previous administrations, including both Trump and Biden, have limited access to other critical technologies like semiconductors and supercomputers, but this move raises new constitutional and commercial concerns. The decision follows broader debates about controlling AI development and distribution, particularly regarding national security risks. The order applies to foreign nationals globally, meaning individuals outside the U.S. with access to Anthropic’s models will no longer be able to use them. The company’s statement did not provide further details on the scope of the restrictions or potential exceptions. The move signals a potential shift in how the U.S. regulates advanced AI technologies, especially those with potential military or geopolitical implications. Anthropic’s compliance reflects the growing tension between innovation and security in the AI sector. While the company has not publicly opposed the order, the decision could impact its international partnerships and research collaborations. The Commerce Department’s action underscores the U.S. government’s willingness to intervene directly in AI development to mitigate perceived risks. This development comes as the U.S. continues to monitor AI advancements closely, particularly those that could be exploited by adversarial foreign actors. The restrictions may also set a precedent for future regulations on AI models, prompting other companies in the sector to prepare for similar measures. The full implications of this order remain unclear, but it signals a significant escalation in AI governance.