Court Upheld Pentagon Ban on Anthropic

The D.C. Circuit Court of Appeals ruled that the Department of Defense can label Anthropic a national security risk.

Updated on Sept. 25, 2026 in Artificial Intelligence

Live Poll

Should the federal government be allowed to blacklist private AI companies over contract disagreements?

The D.C. Circuit Court of Appeals ruled 2-1 to uphold the Pentagon's decision to blacklist Anthropic. The court rejected the artificial intelligence firm's claims that the exclusion violated its First and Fifth Amendment rights.

Why it matters

The Pentagon blacklisted Anthropic after negotiations stalled over the company's refusal to allow its Claude AI models to be used for mass surveillance or fully autonomous lethal weapons. Officials argued these restrictions hindered the military's ability to perform required tasks.

The D.C. Circuit Court of Appeals issued a 2-1 ruling in favor of the Department of Defense. This decision upheld the Pentagon's designation of the firm as a national security risk under the Federal Acquisition Supply Chain Security Act.

The players

Anthropic

Anthropic is an artificial intelligence research company that produces the Claude family of large language models.

Department of Defense

The Department of Defense is the United States federal executive department responsible for coordinating and supervising all agencies and functions of the government related directly to national security.

Karen LeCraft Henderson

Karen LeCraft Henderson is a judge who served as the dissenter in the D.C. Circuit Court of Appeals ruling regarding Anthropic.

The details

The court determined the Pentagon's exclusion was based on a refused contract term regarding model usage rather than an attempt to suppress the company's views on AI regulation. This ruling contradicts an earlier decision from a California federal judge who had found the government violated the First Amendment.

Earlier this year, negotiations between Anthropic and the Pentagon fell apart.
Anthropic sued the Trump administration in August 2026.
A California judge ruled in August 2026 that the government violated the First Amendment.
The D.C. Circuit Court of Appeals issued its ruling on September 18, 2026.

The Tech Race

This case illustrates the friction between private AI developers and national security requirements as the government attempts to integrate advanced software into military operations. The ruling clarifies the legal threshold for the federal government to restrict contractors under the Federal Acquisition Supply Chain Security Act.

This decision impacts how private tech firms interact with government contracts and potential restrictions on their AI tools. Developers and users may see stricter requirements for software compliance as the Pentagon defines acceptable use standards for sensitive technology.

The takeaway

The court's decision establishes that contractors may face exclusion if they refuse to align their AI model capabilities with government operational needs. This precedent highlights the growing complexity of balancing corporate safety mandates with federal security mandates.

What happens next

Anthropic is currently considering whether to seek further review of the appellate court decision, which may eventually lead to a Supreme Court case due to conflicting rulings from different jurisdictions.