A federal judge has struck down the Pentagon’s designation of Anthropic as a “supply chain risk,” ruling that the label was illegal retaliation against the AI company for refusing to let the military use its Claude models without restriction.
U.S. District Judge Rita Lin issued a 59-page order late Thursday in the U.S. District Court for the Northern District of California, finding that the Department of War’s action violated Anthropic’s First Amendment rights and its Fifth Amendment right to due process. Lin called the designation “illegal and baseless” and “arbitrary and capricious,” writing that the government’s evidence of an actual security risk was “slim.”
A dispute over military limits
The fight began in February, when President Trump and then-Defense Secretary Pete Hegseth accused Anthropic of endangering national security. At the center of the disagreement was how far the Pentagon could push Claude’s use: officials wanted the models available for “all lawful purposes,” including sensitive military and intelligence applications, while Anthropic held to two limits it has repeatedly defended in public — no fully autonomous weapons, and no mass domestic surveillance of Americans.
On March 4, the Department of War formally notified Anthropic that it had been labeled a supply-chain risk, a designation that can bar a company’s products from federal contracts on national-security grounds. The Pentagon has other blacklisting tools too — its 1260H list targets Chinese military-linked firms — but Anthropic’s case shows the same “risk” framework can also be turned against a domestic critic. Anthropic sued weeks later, arguing the label was punishment for its public stance on AI safety rather than a genuine security finding, an argument Lin’s ruling largely accepted.
What changes now
Lin wrote that the true motive behind the designation appeared to be punishing Anthropic for its “arrogance” in criticizing government policy, not a genuine supply-chain concern. An Anthropic spokesperson said the company welcomed the ruling and remains “focused on working productively with the government” on national-security uses of its AI. The government is expected to appeal, and a related Anthropic challenge is still pending before the D.C. Circuit Court of Appeals — part of a wider standoff over how the White House decides which companies get privileged access to frontier AI for federal and defense work, an area now shaped by evolving regulation.
Read also: Anthropic’s own account of the dispute, “Where things stand with the Department of War”