A federal judge in California has ruled that the administration's designation of Anthropic as a supply chain risk was illegal.

US District Judge Rita Lin said in her ruling that the defense secretary's labeling of the company as a risk to national security signified "unlawful retaliation" in violation of the First Amendment, and that the decision was "arbitrary and capricious." Lin also found that Anthropic was denied due process as required under the Fifth Amendment.

Where the dispute came from

Earlier this year the administration labeled Anthropic a supply-chain risk and ordered all federal agencies, even those outside defense, to stop working with the company.

The dispute stemmed from the company holding hard lines on certain safety guardrails. Anthropic refused to open its models in ways that would allow use for fully autonomous weapons and mass surveillance of American citizens. The Pentagon denied it would use the models for anything but lawful purposes, and alleged that Anthropic was trying to control the military's use of models it had bought and paid for.

The judge's reasoning

Lin makes a direct finding about the government's intent: the government's "words and deeds confirm that the challenged actions were based on a desire to make a public example out of Anthropic for its 'arrogance' in criticizing the government."

She also pointed to contradictions within the government's own conduct:

  • The Defense Production Act proposal — the secretary's suggestion of applying that act to Anthropic would mean the company was essential to national security rather than a threat to it.
  • An ongoing contract — the Department of Defense was simultaneously pursuing a contract with the company.
  • Cybersecurity collaboration — the government was working with the company's new model, Mythos, on cybersecurity.

Lin also recorded that Anthropic "undisputedly lacks" any backdoor access to its technology once it hands it over to the DOD.

The most quoted line of the ruling: "Though the Department of War is undisputedly free to select the AI vendor of its choice, the evidence demonstrates that the broad measures imposed on Anthropic were illegal and baseless. The empty invocation of national security is not a blank check to punish and retaliate against government critics."

The company's statement on the ruling struck a measured tone: "We welcome the court's ruling that this supply chain risk designation was unlawful. We remain focused on working productively with the government to harness AI for our national security so all Americans benefit from this technology."

Why it matters

This ruling establishes new ground in how AI companies relate to the state. Until now, a company setting limits in its usage terms counted as a commercial choice; this decision confirms those limits may be protected as speech.

The practical consequence: when a model provider says no to certain military or surveillance uses of its product, excluding it from public procurement in response is no longer an easy move.

There is another side to this, about the commercial cost of drawing a line. When Anthropic set those limits it accepted the risk of losing the entire federal market. The ruling does not remove that risk, but it creates precedent: another company setting a similar limit now has a decision to stand on.

Anthropic filed two separate complaints against the DOD in March — one in California, one in Washington, D.C. The Washington suit is still ongoing, so the matter is not closed.