US Appeals Court Upholds Pentagon's Ban on Anthropic

A divided D.C. Circuit panel ruled 2-1 that the Pentagon may label Anthropic a national-security supply-chain risk and keep its Claude models out of Defence Department systems, reversing an August ruling that called the designation unlawful retaliation.

Key Takeaways

A divided D.C.
Circuit panel ruled 2-1 that the Pentagon may label Anthropic a national-security supply-chain risk and keep its Claude models out of Defence Department systems, reversing an August ruling that called the designation unlawful retaliation.

AI synthesis3 sources0

A federal appeals court in Washington, D.C., ruled 2-1 on Friday that the Pentagon may label Anthropic a supply-chain risk to national security and keep the company's Claude models out of Department of Defense systems, rejecting the AI lab's challenge.

The dispute goes back to the Pentagon's March designation, imposed after Anthropic refused to let its technology be used for autonomous weapons and mass surveillance. Defence Secretary Pete Hegseth argued that the company's safety restrictions could jeopardise military operations. The appeals court found the Pentagon's concerns about battlefield readiness reasonable, and held that Anthropic's free-speech claim fails because the exclusion rests on the company's refusal to accept the Department's terms rather than on punishment for its speech.

The ruling reverses the direction of an August decision: a federal judge in San Francisco had blocked a parallel classification under a different law, calling it unlawful retaliation. Two of the three judges on the D.C. Circuit panel sided with the government; the dissent and Anthropic both argue that the designation is arbitrary and exceeds the Supply Chain Security Act.

Anthropic says it disagrees and is weighing its next steps, including a possible further appeal. The company has said the designation has cost it billions and is complicating a planned IPO, even as US intelligence agencies remain heavy users of its models. The decision widens the executive branch's latitude to exclude contractors on security grounds, and it lands in the middle of a broader argument about how far AI companies may set limits on military use of their systems.