Judge Slaps Hegseth Over Unconstitutional Ban

A federal judge ruled that the Pentagon illegally punished an artificial intelligence company for speaking up in public.

Quick Take

  • U.S. District Judge Rita Lin issued a preliminary injunction on March 26, 2026, blocking enforcement of a Pentagon ban on Anthropic’s Claude AI.
  • Lin called the government’s action “classic illegal First Amendment retaliation” against the company for publicly criticizing the Pentagon’s contract demands.
  • The order covered 17 federal agencies, including the Department of Defense and the National Endowment for the Humanities.
  • A federal appeals court later declined to block the Pentagon’s blacklisting of Anthropic while the underlying lawsuit continues.

A Judge Draws a Hard Line Against Retaliation

Judge Rita Lin of the U.S. District Court for the Northern District of California sided with Anthropic on March 26, 2026. She found the government likely retaliated against the company for going public about a contract fight with the Pentagon. Her order used blunt language, calling the punishment “classic illegal First Amendment retaliation” against a company that dared to criticize the government’s contracting position.

Lin’s injunction barred the Trump administration from enforcing the ban while Anthropic’s lawsuit moves forward. The order did not just target the Department of Defense. It reached 17 named agencies, ranging from the Pentagon to the National Endowment for the Humanities, showing how far the directive had spread.

President Trump had ordered federal agencies on February 27, 2026, to immediately stop using Anthropic’s technology. Defense Secretary Pete Hegseth then designated Anthropic a “supply-chain risk” to national security, a rare and severe label for an American tech company. The government framed the move as protecting national security, not punishing a business partner.

The Fight Started Over Battlefield Guardrails

The dispute began when Hegseth pushed Anthropic to let the Pentagon use Claude “as it sees fit,” including for autonomous weapons and surveillance work the company considered too risky. Anthropic refused to strip out its safety guardrails, and negotiations collapsed. Reuters reported the disagreement centered specifically on how much control Anthropic would keep over battlefield and surveillance uses of its own technology.

The government’s stated reasoning shifted as the fight escalated. What started as a contract language dispute turned into a formal supply-chain-risk designation on March 3, 2026, after the guardrail standoff broke down. Defense contractors were then told they could no longer use Anthropic’s AI products at all.

Company Pushes Back, Then Faces a Setback

Anthropic’s leadership argued it could not “accede” to demands that would let the military use Claude without limits, according to the Associated Press. The company sued, arguing the government’s real motive was punishing it for speaking out, not protecting national security. Judge Lin’s injunction gave that argument real legal weight, at least for now.

The legal fight did not end there. On April 8, 2026, a federal appeals court in Washington, D.C., declined to block the Pentagon’s blacklisting of Anthropic while the case continues. That decision does not settle the underlying lawsuit, but it shows the government still has room to enforce its position during the appeal.

This case matters far beyond one company’s contract dispute. Using national-security labels as leverage against a business that refuses government demands should worry anyone who values limited executive power and free speech. A judge finding “classic” retaliation is not a minor technical ruling. It is a warning that government agencies cannot weaponize security designations to silence criticism, even when defense priorities are genuinely at stake.

The Pentagon still has legitimate reasons to demand accountability from AI vendors handling sensitive military work. Nobody disputes that autonomous weapons and surveillance tools need real safeguards. But the proper path runs through clear rules and due process, not punishing a company after it pushes back in public. That distinction is exactly what this litigation will now test in the months ahead.

Sources:

washingtontimes.com, bbc.com, abcnews.com, reuters.com, pbs.org

Featured on Nation Rise

© 2026 Nation Rise. All Rights Reserved.