‹ Back

Anthropic AI Lab formally Labeled Supply Chain Risk by U.S. Pentagon | News

 

Defense Secretary Pete Hegseth designates Anthropic a national security supply-chain risk amid a dispute over military use of its AI systems.

 

 

The U.S. Department of Defense has formally designated the artificial intelligence company Anthropic as a “supply-chain risk to national security,” a move that effectively bars the San Francisco–based firm from defense contracts and commercial activity with U.S. military partners, following an escalating public dispute over how its technology can be used.

 

The designation, announced Friday by Defense Secretary Pete Hegseth on the social media platform X, comes after months of internal negotiations between Defense Department officials and Anthropic leadership over the conditions under which the company’s AI models, including its flagship Claude system, may be accessed and used by the U.S. military.

 

Hegseth’s announcement states that “effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic,” highlighting the department’s intent to sever the startup’s role in government supply chains.

 

Dispute Over Military Access to AI

 

The standoff stems from the Pentagon’s insistence that Anthropic grant the military unrestricted access to its AI models for all lawful purposes, while the company pushed back on certain uses it views as inconsistent with its safety policies.

 

Anthropic, co-founded by CEO Dario Amodei, had requested assurances that its AI would not be used for mass domestic surveillance of U.S. citizens or fully autonomous weapons systems without meaningful human oversight, in line with its published usage policies.

 

Defense officials countered that current law and Pentagon policies already prohibit such uses and that broad access was necessary to ensure operational flexibility across military applications. A senior Pentagon technology official told CBS News the department had offered compromise language acknowledging existing legal safeguards, but differences over contractual terms persisted.

 

Presidential Directive and Blacklist

 

The dispute took on new urgency when President Donald Trump issued a directive ordering all federal agencies to stop using Anthropic’s technology, citing concerns that the company’s refusal to yield access posed a national security risk. Trump’s announcement gave certain agencies, including the Department of Defense, a six-month phase-out period to transition away from Anthropic’s products.

 

Trump’s directive, posted on Truth Social, asserted that federal agencies no longer needed or wanted Anthropic’s technology, and warned of “civil and criminal consequences” if the company did not assist with the transition.

 

Following the presidential order, Hegseth fulfilled his earlier public pledge to designate Anthropic as a supply-chain risk, applying a status typically used for foreign adversaries to a leading U.S. AI developer.

 

Contractual and Market Impact

 

Prior to the designation, Anthropic had secured a Pentagon contract valued at up to $200 million, under which its AI systems, including Claude, were deployed to advance U.S. national security capabilities. Claude’s integration with government systems included classified networks, and the model was used in various intelligence and defense contexts, including analysis tasks linked to sensitive U.S. operations.

 

Industry analysts and policymakers have indicated that the supply-chain risk designation could disrupt partnerships between the Pentagon and major defense contractors that rely on Anthropic’s technology, as well as broader commercial relationships.

 

Broader Context in AI Policy

 

The confrontation reflects wider tensions between AI developers and government agencies over the balance between ethical safeguards and the demands of national security. Anthropic’s stance on usage restrictions distinguishes its approach from some competitors that have accepted broader access terms for defense purposes.

 

Critics of the U.S. government’s actions, including some lawmakers and technology leaders, have expressed concern that the dispute may deter collaboration between federal agencies and private AI firms or shape industry perceptions of regulatory risk in national security markets.

 

Anthropic’s refusal to accede to unrestricted military access, while consistent with its stated safety principles, has now resulted in a significant escalation with one of its largest potential government customers, marking an unprecedented moment in the relationship between AI developers and the U.S. defense establishment.