US blocks foreign access to Anthropic’s advanced AI models

The U.S. government has ordered artificial intelligence company Anthropic to cut off foreign access to its most advanced AI models, citing national security concerns and a potential security vulnerability. The unprecedented directive, issued by the Department of Commerce, forced Anthropic to disable its powerful Fable 5 and Mythos 5 models for all non-U.S. nationals on June 12.

The move effectively treats frontier AI models as controlled strategic assets, similar to advanced military technology. Commerce Secretary Howard Lutnick stated that these models are not ordinary products for overseas distribution. The action followed reports of a high-risk "jailbreak" vulnerability that could allow users to bypass the models' built-in safety guardrails. Anthropic CEO Dario Amodei had previously disclosed the vulnerability, which escalated into a national security discussion involving senior administration officials and tech leaders.

The sudden restriction has sent shockwaves through the global technology industry, prompting urgent reassessments of dependence on U.S.-controlled AI. European Union officials said they are assessing the practical implications of the U.S. directive, noting...