Judge Rules Trump Administration Illegally Retaliated Against Anthropic
A federal judge has ruled that the US Department of Defense acted unlawfully when it designated AI company Anthropic a supply chain risk, calling the move “illegal and baseless” in an Anthropic ruling that could reshape how the government handles disputes with tech firms over military AI use.
The ruling stems from a lawsuit Anthropic filed after Defense Secretary Pete Hegseth’s department issued the designation, which the company argued exceeded his legal authority. The dispute traces back to Anthropic’s refusal to let the military use its AI models for tasks like surveillance or autonomous weapons systems.
What the Judge Said in the Anthropic Ruling
US District Judge Rita Lin wrote in her ruling that citing national security concerns doesn’t give the government “a blank check to punish and retaliate against government critics.” The Pentagon had ordered companies doing business with the US military to boycott Anthropic entirely, an order Lin found unlawful.
The February designation marked a significant escalation, since supply-chain risk labels are typically reserved for companies based in countries considered threats to US national security. It was the first time an American company had received this kind of public designation from the Defense Department.
Why Anthropic Pushed Back
Anthropic’s lawsuit described the government’s actions as “unprecedented and unlawful.” The company maintained that its AI models weren’t reliable enough to be safely deployed in military weapons systems, and separately opposed using its technology for domestic surveillance, arguing that use would violate individual rights.
The Pentagon countered at the time that private companies shouldn’t be able to constrain military decision-making through their own usage policies. Officials also pointed to Anthropic’s refusal to accept new contract terms as a reason for broader concern about how the company might use its technology going forward.
Political Rhetoric Escalated the Dispute
The conflict took on a sharper political dimension when the White House described Anthropic as “a radical left, woke company attempting to control military activity,” arguing the military answers to the US Constitution, “not any woke AI company’s terms of service.” That framing became a central point of contention throughout the legal proceedings.
Anthropic’s lawsuit argued the company’s business had been directly harmed by the government’s actions and that its right to free speech had been violated in the process.
A Pattern the Judge Flagged Earlier
This wasn’t the first time Judge Lin raised concerns about the government’s motives. During an earlier phase of the lawsuit, she said the administration appeared to be trying to “cripple” Anthropic and “chill public debate” over how AI should be used in military contexts.
“This appears to be classic First Amendment retaliation,” Lin said at the time, a characterization that foreshadowed her latest ruling against the Defense Department.
What Happens Next
Anthropic welcomed the ruling, with a spokesperson telling the BBC the company remains focused on “working productively with the government to harness AI for our national security so all Americans benefit from this technology.” The BBC has contacted both the Pentagon and the White House for comment on the decision.
The case highlights a broader tension shaping how Washington regulates artificial intelligence: balancing national security interests against the free speech and business rights of companies developing the technology. With AI increasingly central to defense policy debates, this ruling could set an important precedent for how far the government can go in pressuring companies that decline specific military contracts or use cases.
Legal experts following the Anthropic ruling say decisions like this could make federal agencies more cautious about using national security justifications in future disputes with private companies, particularly in the fast-evolving AI sector where government and industry interests frequently overlap and sometimes collide.