The US government has ordered artificial intelligence company Anthropic to immediately block foreign access to two of its most advanced AI models, citing national security concerns. The company complied but criticized the lack of transparency from regulators.
The Order and Compliance
On Friday evening, the government directed Anthropic to cut off access to Fable 5 and Mythos 5 for all foreign nations, including those outside the US. The order also covered employees of Anthropic. To meet these demands, the company completely shut down access for all customers, effectively ending global usage of the models overnight.
Anthropic confirmed its compliance in a public statement, noting that it had no choice under the directive. The company emphasized that it values national security but expressed frustration with the process.
Government's Lack of Detail
Anthropic said the government did not provide specific details about the national security concern behind the order. According to the company, any evidence of potential jailbreak risks was shared only verbally, and the vulnerabilities described were minor and already available through other models. Anthropic argued that the lack of written documentation makes it difficult to assess the true threat or to address any legitimate risks.
The episode highlights a growing tension between rapid AI development and government oversight. As models become more capable, regulators are increasingly stepping in to limit their spread, often with limited public explanation.
Why This Matters
This order sets a major precedent for how the US government can intervene in AI deployment. Companies like Anthropic now face uncertainty about when and why their models might be restricted. For users and researchers worldwide, access to cutting-edge AI tools could become subject to sudden geopolitical shifts. The lack of transparency also raises questions about accountability because if the government can shut down AI systems without detailed justification, the entire industry may operate under a cloud of unpredictability.
The immediate impact is felt by developers and businesses that relied on Fable 5 and Mythos 5 for their work. They now must pivot to alternative models, potentially losing time and investment. Broader implications include a chilling effect on cross-border AI research and collaboration, as foreign entities may hesitate to depend on US-made AI tools.
Anthropic has stated it will continue to engage with regulators to seek clearer guidelines. For now, the company remains compliant, but the dispute underscores the urgent need for a more structured framework around AI export controls.
