Federal judge rules Pentagon's designation of Anthropic as illegal

U.S. District Judge Rita Lin issued a written order Thursday finding that the Department of Defense acted outside its legal authority when it labeled artificial-intelligence company Anthropic a supply-chain risk. The judge said the Pentagon’s move was motivated by a desire to make an example of Anthropic for its public criticism of the department’s AI policies, rather than any concrete evidence that the firm would sabotage its models.

Anthropic, which sued the Pentagon in March over the designation, welcomed the ruling. The company said it remains focused on collaborating with the government to apply AI to national-security needs.

In a 59-page opinion, Lin, a Biden-appointed judge, wrote that neither the Constitution nor the statute cited by the Pentagon permits “sweeping penalties” based primarily on the company’s critique of administration views. She previously blocked the Pentagon from enforcing a directive that barred federal agencies from using Anthropic’s chatbot Claude.

Department of Justice attorneys argued that the size and opacity of AI models make them difficult for the Defense Department to evaluate like traditional hardware. Anthropic’s lawyer contended that the government’s actions “profoundly harm Anthropic” and risk chilling speech on a critical issue.

The ruling is expected to be appealed, and the Pentagon has indicated it will contest the decision. Separate litigation concerning another Pentagon rule for declaring Anthropic a supply-chain risk remains pending in a federal appeals court.