Pentagon's AI Supply Chain Label Stands: Court Sides with Government Over Anthropic in Critical National Security Ruling

A three-judge panel from the District of Columbia Court of Appeals has dealt Anthropic a significant legal blow, rejecting the AI firm's emergency request to pause the Pentagon's unprecedented "supply chain risk" designation. The decision, handed down Wednesday, marks a critical moment in the intersection of crypto/AI tech governance and national security policy—with potential ripple effects across the entire tech sector.

The Ruling: Government Wins Round One

The DC Circuit judges flatly denied Anthropic's motion for a stay, determining that national security interests outweigh the company's financial and reputational concerns. "In our view, the equitable balance here cuts in favor of the government," the three-judge panel wrote in their order. Their reasoning was blunt: protecting how the Department of Defense secures AI technology during active military conflict takes precedence over harm to a single private company.

This designation represents uncharted legal territory. It's the first time the US government has ever applied a "supply chain risk to national security" label to an American company. The ramifications extend beyond Anthropic alone—it restricts Pentagon contractors from using Claude, Anthropic's large language model, in their operations.

The Underlying Dispute: Claude, Lethal Weapons, and Presidential Orders

The conflict traces back to July 2025, when Anthropic and the Pentagon inked a deal to make Claude the first LLM approved for classified networks. That partnership collapsed in February when negotiations broke down over military usage restrictions. The government demanded unrestricted access to Claude for military applications; Anthropic refused, citing concerns about lethal autonomous weapons and mass domestic surveillance capabilities.

The situation escalated rapidly. In late February, President Trump ordered all federal agencies to cease using Anthropic products, accusing the company of attempting to "strong-arm the Department of War." Anthropic fired back with a lawsuit in March, framing the campaign as unlawful retaliation. A California district court initially sided with the AI firm, issuing a preliminary injunction in late March and calling Trump's directive "Orwellian."

But here's where federal procurement law creates a Byzantine maze: Anthropic had to challenge the designation on two separate legal tracks simultaneously—constitutional grounds in California federal court and directly at the DC Circuit under the specific statute authorizing the designation. That dual-track requirement explains why Wednesday's DC Circuit loss doesn't necessarily end the fight.

What's Next for the Broader Tech Landscape

Acting US Attorney General Todd Blanche hailed the ruling as a "resounding victory for military readiness," emphasizing that "military authority and operational control belong to the Commander-in-Chief and Department of War, not a tech company."

The court acknowledged that Anthropic will "likely suffer some degree of irreparable harm absent a stay" and recommended "substantial expedition" for the remaining legal proceedings. Translation: this case moves fast.