A US federal judge has ruled that the Pentagon unlawfully blacklisted artificial intelligence developer Anthropic as a national-security supply-chain risk, handing the maker of the Claude chatbot a major victory in its dispute with the Trump administration. US District Judge Rita Lin issued the 59-page order in San Francisco on Thursday, blocking the Defense Department from enforcing the designation and related restrictions.
Lin found that the government retaliated against Anthropic for protected criticism of military AI policy, in violation of the First Amendment. She also concluded that the company received no constitutionally required opportunity to respond before the designation took effect, violating Fifth Amendment due-process protections. The judge further found that the decision exceeded the governing procurement law and was arbitrary and capricious.
The conflict began in February after Anthropic and the Pentagon failed to agree on permitted uses of Claude. Chief executive Dario Amodei resisted removing safeguards intended to prevent the technology from supporting mass surveillance of Americans or fully autonomous weapons. Defense Secretary Pete Hegseth then labeled the company a supply-chain risk and directed military contractors, suppliers and partners to stop commercial activity with it.
Anthropic argued that the measure threatened billions of dollars in lost business and lasting reputational damage, while conflicting with the military's earlier positive assessment of Claude. Justice Department lawyers said the company's limits could create uncertainty about battlefield use and could risk software becoming unavailable during an operation. Lin found no factual basis to believe Anthropic would sabotage its model and said national-security claims did not permit retaliation against a government critic.
The court had temporarily frozen the restrictions in March while the case proceeded. Thursday's ruling now prevents enforcement of Hegseth's directive and rejects the legal foundation used to extend the ban beyond direct Pentagon purchases to companies doing business with the military. The decision could influence how US agencies manage AI suppliers when corporate safety rules conflict with government demands.
Anthropic said it welcomed the decision and remained willing to work with the government on national-security uses of AI. The government is expected to appeal, and a separate, narrower case connected to the dispute remains pending in a federal appeals court in Washington. The next proceedings will determine whether the ruling endures and how far the Pentagon may go in setting conditions for commercial AI systems used in defense.
Comments