What happened
The U.S. Court of Appeals for the D.C. Circuit ruled 2-1 on September 25 that the Department of Defense acted within its authority to label Anthropic a supply-chain risk. This decision upholds the Pentagon's exclusion of Anthropic's Claude AI model from military and defense contractor environments, a move stemming from a failed 2025 contract negotiation.
The D.C. Circuit's 2-1 decision on September 25 affirmed the Pentagon's authority to classify Anthropic as a supply-chain risk. The court determined that the Department of Defense had sufficient grounds to make this designation under federal supply-chain law, specifically noting that the government is not required to prove that a company intends to cause harm. Instead, the court found that the company's refusal to allow Claude to be used for all lawful purposes—including autonomous weapons and domestic surveillance—constituted a sufficient basis for the risk designation.
The conflict originated from a $200 million contract signed in July 2025. Negotiations to integrate Claude into the 'GenAI.mil' environment stalled when the Pentagon demanded unrestricted access for all lawful uses. Anthropic refused to remove its safety-based restrictions on autonomous weaponry and mass surveillance. Following a February 27 deadline for compliance, the Pentagon officially designated the company a risk on March 3 and subsequently signed AI agreements with seven other technology firms.
The court has stayed the effect of the ruling, allowing Anthropic to pursue further legal avenues, including a rehearing or a Supreme Court appeal. This ruling stands in contrast to a separate decision by a federal judge in San Francisco, who previously ruled that a different Pentagon designation against Anthropic violated the First Amendment.
Why it matters
This ruling establishes a significant legal precedent regarding the government's power to regulate AI providers based on their refusal to comply with specific deployment requirements. By affirming that the Pentagon does not need to prove malicious intent to designate a company as a supply-chain risk, the court has granted the military broad discretion to prioritize its operational requirements over the safety or ethical restrictions imposed by AI developers. This creates a clear conflict between private policies and federal defense procurement, potentially forcing other AI companies to choose between government contracts and their own internal usage guidelines.
The ruling highlights a critical tension between the federal government's need for unrestricted AI capabilities in defense and the safety-oriented constraints often embedded in commercial AI models. By prioritizing the Pentagon's operational requirements, the court has signaled that companies may be excluded from the defense industrial base if their safety prevent the military from utilizing the technology as it deems necessary.
This decision sets a high bar for AI companies attempting to challenge government procurement decisions on the basis of their own safety policies. It effectively grants the Department of Defense the power to dictate the terms of AI usage, regardless of the developer's stated ethical or safety objectives, provided the government can frame the restriction as a supply-chain risk.
Interactive Mechanism: How It Actually Works
Explore the underlying technology behind this development interactively.
crm_get_transaction(id='4092').Why can ethical evaluation not be reduced to one model score?
What to watch next
Anthropic retains the right to request a rehearing, seek an en banc review by the full D.C. Circuit, or petition the Supreme Court. The industry will be watching to see if Anthropic pursues these appeals or if the ruling forces a shift in how AI companies negotiate safety-related usage restrictions with federal agencies. Additionally, the existence of a conflicting ruling from a federal judge in San Francisco regarding a separate Pentagon designation suggests that legal uncertainty remains regarding the intersection of policies and First Amendment protections.
The primary focus is on Anthropic's next legal move. The company has the option to request a rehearing, seek a full-court review, or appeal to the Supreme Court. Any further litigation will likely clarify the extent to which the government can override private policies.
The discrepancy between this D.C. Circuit ruling and the earlier San Francisco federal court decision creates a split in legal interpretation that may eventually necessitate higher-level judicial intervention to resolve the conflict between national security procurement and First Amendment rights.