In a major win for the San Francisco-based AI company Anthropic, a federal judge has temporarily blocked the Pentagon from labeling the company as a supply chain risk. U.S. District Judge Rita Lin ruled on Thursday that the Trump administration’s actions against Anthropic appeared arbitrary, capricious, and potentially devastating to the company’s future.

The drama started when negotiations between Anthropic and the Pentagon over a defense contract fell apart. The company wanted to prevent its popular Claude chatbot from being used in fully autonomous weapons or surveillance of Americans. Instead of accepting Anthropic’s ethical stance, the Pentagon and Defense Secretary Pete Hegseth responded with what the judge called “broad punitive measures”. This included labeling Anthropic as a supply chain risk and enforcing a presidential directive ordering all federal agencies to stop using Claude.

What made this ruling particularly significant is Lin’s language around government accountability. She wrote that “nothing in the governing statute supports the Orwellian notion that an American company may be branded a potential adversary and saboteur of the U.S. for expressing disagreement with the government”. Basically, she called out the administration for retaliating against a company simply for having ethical boundaries.

During a 90-minute hearing in San Francisco federal court, Lin questioned why the Pentagon took such an extreme step, using a rare military authority that’s typically reserved for foreign enemies, just because contract negotiations didn’t work out. She noted that if the Pentagon was genuinely concerned about Claude’s integrity, they could simply stop using it. “Instead, these measures appear designed to punish Anthropic”, Lin wrote in her decision.

Anthropic responded with a statement saying they were “grateful to the court for moving swiftly” and confident they’ll win on the merits. The company emphasized that while they’re fighting for their survival, they still want to work with the government on AI safety.

The company also has another case pending at the federal appeals court in Washington, D.C., involving a different Pentagon rule attempting to label Anthropic as a supply chain risk. Lin’s order takes effect in a week and doesn’t require the Pentagon to keep using Anthropic’s products or prevent them from switching to other AI providers.

What’s wild is that a bunch of major players filed legal briefs supporting Anthropic, including Microsoft, tech industry groups, retired military leaders, and even a group of Catholic theologians. That kind of coalition suggests people across different sectors are concerned about the government punishing companies for taking ethical stands on AI development.

This case is bigger than just one company versus the Pentagon. It’s about whether the government can crush a business simply because it disagrees with how its technology should be used. For now, at least, the courts are saying that’s not okay.