The Trump administration must lift its ban on Anthropic PBC’s artificial intelligence technology for federal agencies, a US judge ruled.

US District Judge Rita F Lin in San Francisco sided late Thursday with the Claude chatbot maker in its challenge to the Department of Defense, finding that the ban isn’t adequately justified.

The judge wrote that while the government “is certainly owed deference on weighty issues of national security,” its “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, not based on any articulable basis to believe that Anthropic would actually sabotage its model.”

The ruling is a key victory for the AI giant, which has been locked in a high-stakes dispute with the administration since negotiations with the Pentagon over the use of the company’s AI technology by the military fell apart earlier this year.

Anthropic said it welcomed the court’s ruling that the 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,” the company said in a statement.

The White House and the Justice Department didn’t immediately respond to requests for comment.

RELATED ARTICLES

• Trump says he doesn’t see Anthropic as US security threat
• Anthropic sues US govt over supply chain risk label
• Anthropic shuts down Mythos access after sweeping US order
• Anthropic accidentally releases source code for Claude AI agent

Anthropic had sought assurances that its technology would not be used for mass surveillance of Americans or autonomous weapons deployment, but the government resisted any restrictions on its use due to national security concerns.

After discussions broke down, the Department of Defense labeled Anthropic a risk to the US supply chain, blocking government agencies from using the company’s technology.

Anthropic sued, accusing the government of unlawfully retaliating against the company. Lawyers for the chatbot maker have said that a ban would result in billions of dollars in lost revenue.

The ban has been blocked from taking effect since March, under a separate order from Lin. But a lawyer for the government said in late July that the Pentagon has continued to “wind down” its use of Anthropic products and intended to complete that process by Sept. 30.

In her ruling, District Judge Lin made clear that the Defense Department isn’t prevented from taking any lawful action that was available to it before it set the supply chain designation.

“For example, this order does not require the Department of War to use Anthropic’s products or services and does not prevent the Department of War from transitioning to other artificial intelligence providers, so long as those actions are consistent with applicable regulations, statutes, and constitutional provisions,” the judge wrote.

The California case is one of two legal challenges brought by Anthropic over being labeled as a threat. The company also sued in Washington’s federal appeals court, using a different legal argument.

A three-judge panel of that court has yet to issue a ruling, but expressed skepticism during a May hearing that Anthropic could legally second-guess the government’s assessment that it posed a risk to the supply chain.

The case is Anthropic v. US Department of War, 26-cv-01996, US District Court, Northern District of California (San Francisco).