The White House is drafting guidance that would enable federal agencies to bypass Anthropic's supply chain risk designation and deploy new AI models, including its most powerful, Mythos, according to sources familiar with the matter. The move represents a stark reversal for the Trump administration, which had previously labeled the company an unacceptable security risk requiring removal from government systems.
The draft executive action under development could provide a mechanism to de-escalate tensions with Anthropic, with one source describing the effort as a way to "save face and bring em back in." Earlier this month, White House Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent held a productive introductory meeting with Anthropic CEO Dario Amodei.
The administration is convening companies across various sectors this week to inform the potential executive action and establish best practices for deploying Mythos, including "table reads" of possible guidance. The White House push comes as Congress also engages with AI firms, with OpenAI and Anthropic recently briefing House Homeland Security Committee staff on their cyber-capable models in separate classified sessions.
Anthropic has held off on a public release of Mythos Preview due to its ability to quickly find and exploit critical security flaws. The company's cautious release strategy contrasts with OpenAI's tiered approach for its GPT-5.4-Cyber model. Both firms are working with federal agencies to provide access.
Critics argue that reversing the ban risks exposing sensitive government systems to models with untested security vulnerabilities, though the administration maintains it can manage those risks through the planned guidance. The White House declined to comment on the draft executive action's timeline.