Anthropic wins the California case

A federal judge in California has ruled that the Pentagon's decision to label Anthropic a supply-chain risk was unlawful, giving the Claude maker its first court victory in a dispute that has hung over its government business since March.

U.S. District Judge Rita Lin found that the designation amounted to unlawful retaliation under the First Amendment, was arbitrary and capricious, and denied Anthropic the pre-deprivation process required by the Fifth Amendment. CNN reported that the court ordered the designation removed Thursday evening.

The label had blocked the Defense Department and its contractors from using Anthropic products. TechCrunch reported that the broader government response also pushed federal agencies outside defense to stop working with the company, while a separate Anthropic lawsuit over the same dispute continues in Washington, D.C.

The case grew out of Anthropic's refusal to loosen safety limits for certain military uses, including fully autonomous weapons and mass surveillance of U.S. citizens. The Pentagon has said it sought lawful access to AI tools and objected to a private company setting limits on military use.

The ruling does not settle every government-contracting question around Anthropic, but it narrows the immediate blacklist fight. For AI labs selling into national-security work, the decision is a notable boundary: procurement discretion remains broad, but courts may scrutinize labels that punish a vendor for public policy positions.