Reported by Simon Daniel Yusuph l Journalist at Weng Global
A US federal appeals court has upheld the Pentagon’s decision to designate artificial intelligence company Anthropic as a national security supply-chain risk, allowing restrictions that exclude the company from military contracting to remain in place.
The ruling by the US Court of Appeals for the District of Columbia Circuit on Friday, September 25, 2026, came after Anthropic challenged the Pentagon’s decision to restrict its access to US military contracts.
The 2-1 decision represents a setback for Anthropic in its continuing legal dispute with the US government over how its artificial intelligence systems may be used by the military.
The case centres on Anthropic’s refusal to remove safeguards governing certain uses of its Claude AI models, including restrictions relating to fully autonomous weapons and mass surveillance.
Court Upholds Pentagon’s Supply-Chain Designation
The appeals court backed the Pentagon’s authority to designate Anthropic as a supply-chain risk, leaving the company excluded from defence contracts and related procurement arrangements.
The ruling followed an earlier decision in August by US District Judge Rita Lin, who found that the government’s actions against Anthropic were unlawful.
The August ruling had concluded that the Pentagon’s designation did not satisfy the statutory requirements governing supply-chain risks and that several government actions violated Anthropic’s constitutional protections.
Court records show that the earlier case involved a dispute over orders issued by Defence Secretary Pete Hegseth directing government agencies to stop using Anthropic’s technology and designating the company as a supply-chain risk.
The district court found that the government had exceeded its legal authority in parts of its response to Anthropic and that the record did not establish the type of security threat contemplated by the relevant law.
The latest appeals court ruling, however, has restored the Pentagon’s position on the central supply-chain designation.
Dispute Began Over AI Safety Restrictions
The dispute escalated earlier in 2026 after Anthropic and the Pentagon disagreed over restrictions on the military use of the company’s artificial intelligence technology.
Anthropic had sought limits on the use of its systems for fully autonomous weapons and mass domestic surveillance.
The Pentagon objected to restrictions that it considered capable of limiting the military’s ability to use AI for lawful purposes.
Following the disagreement, the administration directed federal agencies to stop using Anthropic’s technology and moved to prevent defence contractors and suppliers from doing business with the company.
The Pentagon subsequently designated Anthropic a supply-chain risk under federal law.
The designation had significant consequences because it prevented Anthropic from participating in certain defence procurement arrangements and also affected companies working with the US military.
Earlier Court Ruling Favoured Anthropic
The latest appeals court decision follows a sharply different ruling from the Northern District of California.
On August 27, Judge Rita Lin ruled in Anthropic’s favour on several major claims, finding that the government’s actions violated the company’s First Amendment and due-process rights and that the Pentagon’s supply-chain designation was contrary to the governing statute.
The court also found that the government had not demonstrated that Anthropic’s conduct met the statutory definition of a supply-chain risk.
According to the court record, the relevant law is concerned with risks such as an adversary sabotaging systems, introducing malicious functions or otherwise subverting covered systems.
The district court concluded that Anthropic’s public position on restrictions governing its AI products did not fall within that definition.
Reuters reported at the time that Judge Lin described the Pentagon’s blacklisting action as unlawful retaliation and found constitutional and statutory problems with the government’s conduct.
The government appealed aspects of the ruling, leading to the latest decision from the D.C. Circuit.
Anthropic’s Position on AI Safety
At the centre of the dispute is a broader question about how advanced AI systems should be deployed in military settings.
Anthropic has maintained restrictions on certain uses of its technology, particularly fully autonomous weapons and mass surveillance.
The company has argued that safeguards are necessary to prevent its models from being used in ways that could create significant risks.
The Pentagon, meanwhile, has sought access to advanced AI capabilities for defence and national-security applications without restrictions that officials believe could interfere with military operations.
The disagreement has become part of a wider debate in the United States over the balance between AI safety measures, commercial interests and national-security requirements.
What the Appeals Court Decision Means
The appeals court ruling leaves the Pentagon’s supply-chain designation in force.
That means Anthropic remains excluded from the relevant military supply chain and cannot participate in defence contracting arrangements affected by the designation.
The decision also illustrates the legal uncertainty surrounding the government’s treatment of major AI companies as the US military increasingly seeks to incorporate artificial intelligence into defence operations.
The case is significant beyond Anthropic because it raises questions about the extent to which government agencies can impose restrictions on technology companies whose products are used in national-security environments.
It also highlights the tension between government demands for unrestricted access to advanced AI systems and companies’ efforts to establish limits on potentially harmful applications.
Legal Dispute May Continue
Anthropic has indicated that it is considering its options following the appeals court ruling.
The company could seek further judicial review, although the precise next step will depend on its legal strategy and the available avenues for appeal.
The appeals court ruling does not erase the August district court decision, but it changes the legal position surrounding the Pentagon’s supply-chain designation.
The separate proceedings in different federal courts also mean the broader dispute over the government’s actions toward Anthropic may continue.
Why the Case Matters for the AI Industry
The Anthropic-Pentagon dispute has implications for the wider artificial intelligence industry because major AI developers are increasingly seeking government contracts while simultaneously establishing safeguards around how their systems can be used.
For governments, advanced AI models are becoming increasingly important for research, intelligence, cybersecurity, logistics and military planning.
For AI companies, government contracts can represent a major commercial opportunity, but restrictions imposed on how their systems are deployed can also become a point of disagreement.
The Anthropic case therefore places a wider question before US policymakers and courts: how should the government balance national-security requirements with legal limits and the conditions technology companies place on the use of their products?
The latest appeals court ruling has not resolved that broader policy debate.
Instead, it has determined, at this stage of the litigation, that the Pentagon’s supply-chain designation of Anthropic can stand.
What Happens Next
Anthropic’s next legal move will determine the immediate future of the dispute.
The company may seek further review of the appeals court decision, while the Pentagon can continue enforcing the supply-chain designation under the latest ruling.
The case is also likely to remain closely watched by other AI companies developing technologies for government and military customers.
For now, the appeals court decision leaves Anthropic facing continued restrictions on its participation in the US defence supply chain, despite its earlier victory in federal district court.
The dispute remains an important test of the legal boundaries surrounding government procurement, AI safety safeguards and the use of advanced artificial intelligence in national-security operations.
Weng Global – stories beyond borders
Sources
- US Court of Appeals for the District of Columbia Circuit
- US District Court for the Northern District of California
- Reuters
- Breaking Defense
- The Next Web
