The Trump administration has moved to block foreign governments, companies, and individuals from accessing Anthropic's most advanced artificial intelligence models, marking a significant escalation in efforts to treat cutting-edge AI as a national security asset. Anthropic, a leading AI developer, now faces a dual designation: deemed too dangerous for government use by the Pentagon and subject to export controls by the Commerce Department.
Commerce Department Takes Action
Commerce Secretary Howard Lutnick issued a letter to Anthropic CEO Dario Amodei on Friday, placing the company's Mythos 5 and Fable 5 AI models under strict export controls. The decision, effective immediately, bars any export, re-export, or domestic transfer of these models without a license. Anthropic will also be required to submit additional applications for individually validated licenses, with non-compliance resulting in financial and civil penalties.
The model needs to remain locked down until the U.S. government's national security apparatus is hardened, an administration official stated.
The move comes after another company claimed to have successfully jailbroken Mythos, raising alarms within the administration about potential national security risks. Despite attempts to persuade Anthropic to pause the release of its latest models, the administration was forced to act.
Broader Context
This decision aligns with the Trump administration's broader focus on AI security. Earlier this month, President Trump issued an executive order requiring pre-deployment testing for the most advanced AI models. However, the order remains voluntary and avoids imposing a licensing regime to prevent stifling innovation.
Anthropic has a partnership with the Center for AI Standards and Innovation at the Commerce Department for pre-deployment testing. Administration officials emphasize that the goal is not to harm the industry but to ensure national security while fostering continued innovation.