WASHINGTON (AP) — A panel of appellate judges appeared to be divided on Tuesday over a legal dispute between the Pentagon and technology company Anthropic, which claims Defense Secretary Pete Hegseth unlawfully and falsely branded it as a national security risk for raising ethical and safety concerns about AI usage in war.

Three judges from the U.S. Court of Appeals for the District of Columbia Circuit didn’t indicate how soon they would rule on Anthropic’s appeal, but some of their questions and remarks hinted at how they might decide the case.

Judge Karen LeCraft Henderson said she sees no evidence to support the Pentagon’s determination that Anthropic poses a supply-chain risk to national security.

“To me, this is just a spectacular overreach by the (Defense) Department,” said Henderson, who was nominated by Republican President George H. W. Bush.

Judge Neomi Rao, who was nominated by Republican President Donald Trump, questioned what basis the court could have for second-guessing Hegseth’s judgment. The Pentagon’s dispute with Anthropic centers on how AI technology can be used in fully autonomous weapons and potential surveillance of Americans.

“I take the secretary to be making more general points than the ones that you’ve identified,” Rao told Anthropic attorney Kelly Dunbar. “It’s about risk, and they say, ‘Well, based on what we know, we can’t trust that the (AI) model may not have something embedded within it that is going to create a problem for military capabilities.”

Anthropic filed lawsuits in Washington, D.C., and San Francisco after the Pentagon designated the San Francisco-based company as a supply chain risk and Trump ordered all federal agencies to stop using its technology. Anthropic claims the Pentagon is illegally retaliating against it by stigmatizing it with a designation meant to protect against sabotage of national security systems by foreign adversaries.

Anthropic says neither lawsuit seeks to force the government to contract with the company. But it claims that it has been irreparably harmed by Hegseth’s supply-chain risk designation.

Earlier this month, the D.C. circuit rejected Anthropic’s request for an order that would have blocked the Pentagon’s actions while the company’s appeal is pending.

In a separate but related case, a federal judge in San Francisco ruled in Anthropic’s favor last month and blocked the Pentagon from labeling Anthropic as a supply-chain risk.

In a court filing ahead of Tuesday’s hearing in Washington, Anthropic said it can’t manipulate its AI tool Claude once it is deployed in classified Pentagon military networks.

But a Justice Department attorney, Sharon Swingle, told the D.C. Circuit judges that Anthropic clearly has the ability to interfere with the Pentagon’s usage of the company’s AI model “for critical military operations.”

“It’s undisputed that the failure of the model in active military operations could have catastrophic national-security consequences and put service members’ lives at risk,” she said.

Dunbar said Hegseth’s designation of Anthropic as a supply-chain risk “defied congressionally mandated procedures, exceeded statutory limits and violated the Constitution.”

“For the first time ever, the secretary turned a powerful national security authority against an American company, and he did so to gain leverage in a contract dispute,” Dunbar said.

Judge Gregory Katsas, another Trump nominee, also heard Tuesday’s arguments.

Michael Kunzelman, The Associated Press