Anthropic Sues Trump Administration After Pentagon Labels It 'Supply-Chain Risk' For Refusing To Weaponize Claude

Anthropic filed lawsuits in two federal courts after the Pentagon formally designated the company a "supply-chain risk" and President Trump ordered federal agencies to stop using its Claude model.

“ADVERTISEMENT Anthropic has filed a lawsuit against the United States Department of Defense in an attempt to block a decision that labels the company a national security supply-chain risk, intensifying a dispute over how artificial intelligence can be used in military operations”

@storyboard18

The complaints seek to undo the Pentagon's decision and block enforcement of related agency directives, framing the dispute as a direct legal challenge to the administration's recent, public pressure on the company.

@storyboard18

The filings name multiple agencies and officials and ask courts to intervene as Anthropic contests an action it says will immediately curtail its government business.

Anthropic’s legal complaints argue the designation was unlawful retaliation that violated constitutional protections and federal procedures.

The company’s filings assert First Amendment retaliation and Fifth Amendment due-process claims.

ABC

They also invoke the Administrative Procedure Act and argue the government exceeded statutory authority by applying a supply-chain law the company says targets foreign threats rather than domestic firms.

Anthropic’s lawyers ask judges to overturn the label and enjoin agencies from enforcing the ban while litigation proceeds.

At the heart of the dispute are the defensive guardrails Anthropic imposed on Claude, specifically limits barring its use for domestic mass surveillance of Americans and for fully autonomous weapons.

“Select “More options” to see additional information, including details about managing your privacy settings”

ABC10

The Pentagon insists on access for "any lawful" use.

Anthropic says it refused demands to remove those restrictions during negotiations, including a February meeting between CEO Dario Amodei and Defense officials, and that the DoD's move followed after those talks stalled.

Reporting also links the designation to classified military deployments of Claude, including alleged use in operations tied to Iran.

The government action has had immediate commercial and operational effects and prompted mixed industry responses.

Several federal agencies and contractors began cutting ties or terminating contracts.

Al Jazeera

Anthropic warns of substantial near-term revenue losses.

Leading cloud vendors said they will continue offering Claude for non-Defense uses while segregating Pentagon work.

Rival AI suppliers and some researchers weighed in publicly.

OpenAI reached terms with the Pentagon in a deal that underscored shifting competitive dynamics in supplying classified and defense-related AI.

Observers and legal experts say the case raises novel precedent questions because the supply-chain designation is normally used against foreign adversaries and commentators describe the step as highly unusual for a U.S. firm.

“Last week, a Pentagon official said the two sides were no longer in active talks”

Al Jazeera

Anthropic and opponents frame the dispute as a defining test of whether private companies can set ethical limits on how governments use advanced AI, and the courts' rulings, including a D.C. Circuit review, will shape future relations between frontier AI labs and national security agencies.

BBC