Anthropic CEO Refuses Pentagon AI Demands, Tensions Escalate

A bitter feud between the Pentagon and artificial intelligence startup Anthropic intensified on Friday after the Trump administration cut off the company’s military contracts. The conflict centers around Anthropic’s refusal to allow its Claude AI model to be used for mass surveillance or autonomous weapons, citing concerns over American values and national security.

In an exclusive interview with CBS News, Anthropic CEO Dario Amodei reaffirmed the company’s desire to work with the military, but insisted it would only do so if its concerns are addressed. “We are still interested in working with them as long as it is in line with our red lines,” Amodei said, referring to the company’s insistence on safeguards against surveillance and the development of autonomous weapons.

The Pentagon, which has deployed Anthropic’s Claude model on its classified networks, demands the right to use it for “all lawful purposes,” claiming that it does not intend to use the technology for surveillance or autonomous weapons. Despite this, Amodei remains firm on the need for explicit guardrails, warning that AI’s potential could outpace current legal frameworks and lead to risks like unlawful surveillance or misfires from autonomous weapons.

The standoff escalated when President Trump ordered federal agencies to halt use of Anthropic’s technology. Defense Secretary Pete Hegseth declared the company a “supply chain risk,” urging military contractors to sever ties with the startup.

Amodei defended his company’s stance, emphasizing that their red lines were non-negotiable, particularly regarding the use of AI in military operations that could threaten civilian safety. “We don’t want to sell something that could get our own people killed or that could get innocent people killed,” he said.

The Pentagon has countered that federal law already restricts mass surveillance and fully autonomous weapons, rendering additional written safeguards unnecessary. In a bid for compromise, Pentagon officials offered to acknowledge existing legal frameworks but Anthropic dismissed this as insufficient, calling it “legalese” that could be bypassed.

As tensions mount, top military figures have accused Anthropic of attempting to impose its values on the government. Hegseth called the company “sanctimonious” and President Trump labeled it a “radical left, woke company” jeopardizing national security.

Despite the dispute, Amodei expressed a desire for a resolution. “For our part and for the sake of U.S. national security, we continue to want to make this work,” he said. However, with no agreement in sight by Friday, the Pentagon is set to phase out Anthropic’s technology within six months and transition to alternative solutions.

Amodei called the Pentagon’s actions “retaliatory and punitive” and vowed to challenge them legally. He emphasized that his company’s opposition stems from a commitment to national security, adding, “Disagreeing with the government is the most American thing in the world.”