Judge Blocks Pentagon’s Anthropic ‘Supply Chain Risk’ Designation

A federal judge in California temporarily blocked the U.S. Department of Defense from enforcing its designation of Anthropic as a national security “supply chain risk,” a move that had threatened to cut the AI company off from federal work and pressure contractors to stop using its technology. The dispute began after President Donald Trump and Defense Secretary Pete Hegseth moved to halt government use of Anthropic tools and formally label the company a risk following Anthropic’s refusal to permit uses tied to mass surveillance of Americans and fully autonomous lethal targeting. Anthropic sued, arguing the government retaliated against protected speech and imposed the designation without notice or a meaningful chance to respond.

In granting a preliminary injunction, the court said Anthropic was likely to succeed on First Amendment, due process, and Administrative Procedure Act claims, finding the government’s stated sabotage and wartime-disablement rationale appeared unsupported and that the relevant law was likely misapplied. The ruling came after public criticism from Sen. Elizabeth Warren, internal Pentagon guidance allowing limited exemptions because Anthropic tools were difficult to remove from some systems, and reports that federal agencies continued testing Anthropic’s cyber-focused Mythos model despite the ban. Subsequent White House discussions signaled the administration may soften its stance even as the Pentagon expanded AI deals with other major tech firms and litigation over Anthropic’s federal status continued.