Washington — Appeals Court Upholds Trump Blacklist on Anthropic AI

A federal appeals court ruled Sept. 25 that the Trump administration had the authority to bar federal agencies and defense contractors from using technology made by Anthropic, the artificial intelligence company behind the Claude chatbot, according to the U.S. Court of Appeals for the District of Columbia Circuit ruling reported by Ars Technica.

The 2-1 decision found the government could add Anthropic to its blacklist even without evidence of malicious intent by the company, resolving one front in a legal fight that began in March, when President Trump and Defense Secretary Pete Hegseth ordered federal agencies to stop using Anthropic's products and prohibited defense contractors from doing business with the company. Anthropic sued in response.

What did the D.C. Circuit actually decide?

The panel wrote that the case "raises profoundly difficult questions about the appropriate military uses of an almost unimaginably powerful new technology." It framed the dispute as a tradeoff between two risks: the government's argument that "overly constrained AI models" could shut down unexpectedly and cause "important military operations to fail," against Anthropic's warning that "unconstrained AI models" could hallucinate "inappropriate targets for lethal military force."

Having laid out that tension, the court said the choice belonged to the executive branch, not the judiciary. Trump and Hegseth "must determine how best to balance the competing risks," the ruling said, adding that Hegseth "did not transgress any limits on his authority under the Supply Chain Security Act or the Constitution." On that basis, the panel denied Anthropic's petitions for review.

The two judges in the majority, Gregory Katsas and Neomi Rao, were both appointed by Trump and served in his first administration — Katsas as a former deputy counsel to the president, Rao in the Office of Management and Budget. The same appeals court had already denied Anthropic's request for an emergency stay in April, so Sept. 25's ruling extends a pattern of the D.C. Circuit declining to block the blacklist while the underlying case played out.

How does this ruling conflict with the California decision?

A federal judge in the U.S. District Court for the Northern District of California reached the opposite conclusion last month, ruling that the blacklisting was illegal because Anthropic did not meet the legal definition of a supply-chain risk. That definition, the district court found, is limited to "the risk that an adversary may sabotage, maliciously introduce unwanted function, or otherwise subvert … a covered system."

The D.C. Circuit's ruling did not dispute that finding directly. Instead, the appeals panel said the two courts were reviewing different legal questions: the district court examined the government's action under one statute, while the D.C. Circuit has exclusive jurisdiction to review the same action under a separate, more permissive legal authority, according to Ars Technica's account of the ruling. That distinction is why two federal courts arrived at opposite outcomes on what is, at bottom, the same dispute.

What is Anthropic saying now?

Anthropic has not conceded the fight. "We respectfully disagree with the court's decision," a company spokesperson told CNBC, as reported by Ars Technica. "Another federal court has already held the government's parallel designation unlawful. We remain confident in our position and are considering all options, including further review."

That leaves Anthropic with two procedural paths: asking the full D.C. Circuit to rehear the case en banc, or petitioning the Supreme Court directly.

The legal fight has not stopped the two sides from talking. Commerce Secretary Howard Lutnick recently said the Trump administration and Anthropic have "patched up" their relationship and are "in tune," according to Ars Technica — a striking contrast to the adversarial posture both sides have taken in court filings.

What happens next in the Anthropic case?

• En banc review: Anthropic can ask all active judges on the D.C. Circuit to rehear the case, which could overturn the 2-1 panel decision.
• Supreme Court petition: Anthropic could bypass further D.C. Circuit review and ask the Supreme Court to take the case directly.
• Conflicting rulings stand for now: The Northern District of California's finding that the blacklist was illegal under the Supply Chain Security Act's malicious-intent standard has not been overturned; it was simply held not to control the broader authority the D.C. Circuit reviewed.
• Contracts remain frozen: Absent a stay or reversal, federal agencies and defense contractors remain barred from using Anthropic's technology while the case proceeds.

Whether the diplomatic thaw described by Lutnick translates into a settlement — or whether Anthropic pushes the case to the Supreme Court — will determine whether the blacklist is a temporary standoff or a lasting precedent for how far the executive branch can go in restricting AI companies over military-readiness concerns.