25 September 2026
Appeals court sides with the Pentagon in its blacklisting of AI company Anthropic
A 2-1 ruling in Washington upheld the Defense Department's label of Anthropic as a supply chain risk, which bars the military and its contractors from using Claude.
A federal appeals court in Washington, D.C., has upheld the Pentagon's decision to treat AI company Anthropic as a national-security supply chain risk. The label stops the US military from using Anthropic's Claude AI models and blocks defense contractors from using them in their work for the Defense Department.
What happened
The three-judge panel split 2-1. Judge Gregory Katsas, joined by Judge Neomi Rao, wrote that the department had ample support for its national-security concerns and that the Defense Secretary stayed within his legal authority. Judge Karen LeCraft Henderson dissented. Katsas and Rao were appointed by President Trump; Henderson was appointed by President George H.W. Bush.
The dispute goes back to talks over how the military could use Claude. Anthropic signed a $200 million Pentagon contract in July 2025, but negotiations over putting Claude on the department's GenAI.mil platform broke down. The department wanted access for all lawful purposes. Anthropic wanted assurances its models would not be used for fully autonomous weapons or domestic mass surveillance. In March, the department issued the supply chain risk label.
The Pentagon relied on two separate designations, so Anthropic had to fight them in two courts. Last month a federal judge in San Francisco ruled one of them illegal. Friday's ruling upheld the other.
Anthropic said it respectfully disagrees and is considering all options, including further review. The court delayed its decision from taking effect so Anthropic can ask for a rehearing, either by the same panel or by the full appeals court. It could also ask the Supreme Court to hear the case.
Why it matters
A supply chain risk label does more than end one contract. It shuts a company out of the military directly, and also out of work done by the many contractors that serve the Pentagon, which is why the label matters well beyond the original $200 million deal.
The case is also a test of how much say an AI developer has over how the government uses its technology. The majority's answer was that decisions about balancing those risks belong to the President and the Defense Secretary. With two courts now reaching different conclusions on the two designations, the fight is not over.
The relationship between Anthropic and the administration has been tense for months, and President Trump has repeatedly criticized Anthropic CEO Dario Amodei on social media.
What to watch
- Whether Anthropic asks for a rehearing by the full D.C. Circuit or goes to the Supreme Court.
- The government's response to the San Francisco ruling against the other designation.
- Whether the Pentagon's ban takes effect once the court's delay runs out.
Based on reporting by CNBC: U.S. appeals court upholds Pentagon designation of Anthropic as supply chain risk. Facts, not advice.
More news
- Anthropic's IPO, in numbers: 825 million shares at $70 to $78, and a $1.4 trillion price tag
- Trump says the Nvidia scandal didn't endanger US security, and stands by Lutnick
- Nvidia's first earnings without Huang: what Kress will be asked on Tuesday
- Buffett gives away $5.5 billion of Berkshire stock in his Thanksgiving letter
- Banks are using AI to trim back offices. Proof that it adds revenue is harder to find
- IMF chief says the world economy leans too hard on AI as circular deals top $1 trillion