In a moment that tests the boundaries between national security and political retribution, a federal judge ruled Thursday that the Pentagon's blacklisting of AI company Anthropic was unlawful retaliation for the company's refusal to subordinate its ethical commitments to military imperatives. Judge Rita Lin found that invoking supply-chain risk statutes designed to guard against foreign sabotage — against an American company, for its speech — violated both the First and Fifth Amendments. The ruling raises a question that will outlast this case: in an age of powerful AI, who holds the moral aut
Federal judge strikes down Pentagon's unlawful blacklisting of AI firm Anthropic
The empty invocation of national security is not a blank check to punish critics
Mark
So the judge said the Pentagon was retaliating against Anthropic for refusing to build weapons. What exactly did Anthropic refuse to do?
Mimi
Anthropic said its Claude AI models weren't reliable enough for autonomous weapons—machines that make kill decisions without human input. They also refused to let the military use Claude for domestic surveillance. Those were hard lines for them.
Luke
Right, but we should be clear: Anthropic's own safety concerns are their stated reason. The Pentagon says it was about contractual terms, not retaliation. The judge sided with Anthropic, but that's a judgment call about intent.
Mark
And this "supply chain risk" designation—why is that such a big deal?
Mimi
It's usually used to block foreign companies from military contracts. Applying it to an American firm for the first time, and doing it publicly, sent a message: comply or be cut off from government work.
Luke
The company said it could cost billions. We don't have independent verification of that number. It's their estimate of potential harm.
Mark
Trump called them "woke" on social media right after they refused. Doesn't that look like retaliation?
Mimi
That's exactly what the judge focused on. The timing, the language, the fact that the Pentagon had praised Claude before—it all pointed to punishment for speaking out.
Luke
The government will argue the social media post doesn't prove the designation was retaliatory. Appeals courts might see it differently. This isn't over.
Mark
What happens now?
Mimi
Anthropic can use its tools with the military again, at least for now. But they're still fighting a second case over civilian government contracts.
Luke
And the government can appeal this ruling. We don't know if it will, or whether a higher court would overturn it.
The Pulse
- The Pentagon's unprecedented designation of Anthropic as a 'supply chain risk' — a label never before applied to an American company — threatened to cost the AI firm billions and set a chilling precedent for any tech company that declines government demands.
- President Trump publicly branded Anthropic 'radical left' and 'woke,' making the political nature of the retaliation difficult to conceal behind the government's national-security justifications.
- Anthropic fought back on constitutional grounds, arguing the punishment targeted its speech on AI safety and denied it any meaningful chance to respond — a dual assault on the First and Fifth Amendments.
- Judge Lin's 59-page ruling was unsparing, declaring that 'the empty invocation of national security is not a blank check to punish and retaliate against government critics,' and immediately lifting the sanctions.
- The victory remains fragile: a government appeal is possible, and a second, separate blacklisting case over civilian contracts still awaits resolution in Washington.
In a moment that tests the boundaries between national security and political retribution, a federal judge ruled Thursday that the Pentagon's blacklisting of AI company Anthropic was unlawful retaliation for the company's refusal to subordinate its ethical commitments to military imperatives. Judge Rita Lin found that invoking supply-chain risk statutes designed to guard against foreign sabotage — against an American company, for its speech — violated both the First and Fifth Amendments. The ruling raises a question that will outlast this case: in an age of powerful AI, who holds the moral authority to set the terms of its use?
On Thursday, a federal judge struck down the Trump administration's effort to blacklist Anthropic, the AI company behind Claude, ruling that the Pentagon had illegally retaliated against the firm for refusing to let the military use its tools for surveillance and autonomous weapons.
The designation, issued in February by Defense Secretary Pete Hegseth, was without precedent — no American company had ever been publicly branded a supply-chain risk under a statute designed to protect military systems from foreign sabotage. Anthropic had declined to modify its AI models for military purposes, citing safety concerns and opposition to domestic surveillance. President Trump responded with public attacks on social media, and Hegseth's move blocked the company from certain military contracts, a blow Anthropic said could cost it billions.
Judge Rita Lin, a Biden appointee, was direct in her 59-page ruling: national security claims, she wrote, are not a blank check to punish government critics. She found the Pentagon's action illegal and baseless, and immediately barred federal agencies from enforcing the order — making permanent a temporary suspension she had granted when Anthropic first sued in March.
The company's lawsuit rested on two constitutional pillars: that the government punished it for its stated positions on AI safety in violation of the First Amendment, and that the Pentagon imposed the designation without any opportunity to respond, violating Fifth Amendment due-process rights. The Justice Department countered that Anthropic's contractual refusals — not its public statements — triggered the action, but the court found that reasoning unconvincing given the timing and the president's visible fury.
The ruling does not end the fight. The government may appeal, and a second designation targeting Anthropic's access to civilian government contracts remains pending in Washington. Still, the decision signals that courts will look hard at whether national-security claims are being used to mask political punishment — and for now, the immediate threat to Anthropic's business has been lifted.
On Thursday, a federal judge dismantled the Trump administration's effort to blacklist Anthropic, the artificial intelligence company behind Claude. Judge Rita Lin ruled that the Pentagon's decision to designate the firm as a "supply chain risk" was illegal retaliation for the company's public refusal to let the military use its AI tools for surveillance and autonomous weapons.
The designation, handed down in February by Defense Secretary Pete Hegseth, was unprecedented. No American company had ever been publicly branded a supply-chain risk under the obscure government-procurement statute meant to protect military systems from foreign sabotage. Anthropic had refused to modify its AI models to serve military purposes, citing safety concerns and opposition to domestic surveillance. President Trump responded by calling the company "radical left" and "woke" on social media. Hegseth's move blocked Anthropic from certain military contracts—a blow the company said could cost it billions in lost business and damage to its reputation.
In her 59-page decision, Judge Lin, a Biden appointee, was direct: "The empty invocation of national security is not a blank check to punish and retaliate against government critics." She found the Pentagon's action "illegal and baseless" and immediately barred federal agencies from enforcing Trump's order to stop using Anthropic's tools. The ruling made permanent a temporary suspension of the sanctions that Lin had ordered in March, when Anthropic first sued.
Anthropics lawsuit rested on two constitutional grounds. The company argued the government violated its First Amendment right to free speech by punishing it for its stated positions on AI safety. It also claimed the Pentagon violated its Fifth Amendment due-process rights by imposing the designation without giving the company a chance to respond or challenge it. The lawsuit noted that the Pentagon had previously praised Claude's capabilities, making the sudden reversal appear driven by politics rather than legitimate security concerns.
The Justice Department's defense was different. Officials argued that Anthropic's refusal to accept the Pentagon's contractual terms—not its public statements—triggered the designation. They claimed the company's restrictions on military use created uncertainty about how the Pentagon could deploy Claude and risked disabling military systems during operations. But the court found this reasoning unconvincing, particularly given the timing and Trump's public fury at the company.
Anthropics victory is not final. The government has the right to appeal, and the company faces a second, separate lawsuit over another Pentagon supply-chain risk designation that could exclude it from civilian government contracts as well. That case remains pending in Washington. Still, Thursday's ruling signals that even in an era of heightened national-security claims, courts will scrutinize whether those claims mask political punishment. For Anthropic, it means the immediate threat to its military business has been lifted—though the broader fight over whether AI companies can set their own ethical boundaries with the Pentagon continues.
Notable Quotes
The empty invocation of national security is not a blank check to punish and retaliate against government critics.— Judge Rita Lin
Anthropic argued the designation could cost the company billions of dollars in lost business and reputational harm.— Anthropic executives
Want the full story?
Read the original at The Guardian ↗
Technology