White House Plans to Revise AI Policy, Reintegrate Anthropic Despite Security Concerns

The White House is developing guidance that would allow federal agencies to bypass Anthropic’s supply chain risk designation and onboard new models, including its most powerful AI model yet, Mythos, according to sources familiar with the matter.

Why This Matters

The Trump administration appears to be reversing its stance on Anthropic, a company it previously deemed a grave security risk, requiring its removal from federal systems.

Behind the Scenes: Draft Executive Action in Progress

A draft executive action, currently in development, could provide the administration with a mechanism to ease tensions with Anthropic. Two sources described the White House’s efforts as an attempt to "save face and bring [Anthropic] back in."

Recent Meetings Signal Shift in Policy

Earlier this month, White House chief of staff Susie Wiles and Treasury Secretary Scott Bessent met with Anthropic CEO Dario Amodei for what both sides described as a productive introductory discussion on potential collaboration between the company and the government.

The White House is also convening meetings with companies across multiple sectors this week to inform the potential executive action and establish best practices for deploying Mythos. These sessions include "table reads" of possible guidance that could walk back the Office of Management and Budget’s (OMB) directive prohibiting the use of Anthropic within the federal government.

Statements from Key Stakeholders

"The White House continues to proactively engage across government and industry to protect our country and the American people, including by working with frontier AI labs."

The collective effort of all involved will ultimately benefit our economy and country. However, any policy announcement will come directly from the President, and anything else is pure speculation."

Anthropic declined to comment on the matter.

Background: The Pentagon and White House’s Evolving Stance

The Pentagon and White House had previously aligned in blacklisting Anthropic, criticizing the company for perceived ideological biases. However, the emergence of Mythos—an AI model capable of automating cyberattacks but also a potential tool for cyber defense—has complicated the situation.

Federal agencies, including the Pentagon, are currently permitted to use Anthropic’s models while legal disputes play out. Notably, the National Security Agency (NSA) is already utilizing Mythos.

Internal Divisions Over Anthropic’s Role

Multiple sources told Axios that while key Pentagon officials remain firmly opposed to Anthropic, other stakeholders view the ongoing conflict as counterproductive and are seeking a resolution.

It remains unclear whether the proposed steps will resolve the Pentagon’s dispute with Anthropic or merely facilitate broader government adoption of its models.

Core Disputes Remain Unresolved

Even if the Pentagon lifts its supply chain risk designation—an outcome some Pentagon officials and Anthropic representatives believe is likely—the underlying conflict persists. Anthropic has refused to sign an agreement allowing the Pentagon unrestricted use of its Claude model, insisting on prohibiting its use for mass domestic surveillance or the development of fully autonomous weapons. The Pentagon has argued that this refusal demonstrates Anthropic’s unwillingness to comply with national security requirements.