Anthropic CEO Defends “Red Lines” on AI Weapons as Pentagon Standoff Escalates

Anthropic CEO Defends "Red Lines" on AI Weapons as Pentagon Standoff Escalates

Anthropic CEO Dario Amodei defended his company's refusal to provide unrestricted AI access to the Pentagon, telling CBS News that while his startup is committed to national defense, it won't budge on prohibitions against domestic mass surveillance and fully autonomous weapons—even as the Trump administration cuts off federal contracts.

The standoff has immediate implications for finance chiefs across the defense industrial base. Defense Secretary Pete Hegseth has designated Anthropic a "supply-chain risk," meaning contractors working on Pentagon projects cannot use the company's AI models—a restriction that could force technology audits and vendor switches at major aerospace and defense firms.

President Trump ordered federal agencies to stop working with Anthropic on February 28, 2026, with the Pentagon receiving a six-month phaseout period. The ban came after negotiations broke down over the company's insistence on maintaining what Amodei calls "red lines" for AI deployment, despite the military demanding unfettered use in all legal scenarios.

Amodei, speaking shortly after the White House order, emphasized that Anthropic was "the first to serve the defense community in a classified setting" and maintains "a substantial public sector team." He told CBS that the company supports 98% to 99% of military use cases and remains open to continued collaboration.

"I believe we have to defend our country from autocratic adversaries like China and like Russia," Amodei said. "We've been very lean forward" on defense work.

But the CEO drew a firm line on two applications. On domestic surveillance, Amodei argued that AI has transformed previously impractical monitoring into a powerful capability that operates within existing legal frameworks—but shouldn't. "That actually isn't illegal. It was just never useful before the era of AI," he explained. "The technology's advancing so fast that it's out of step with the law."

On autonomous weapons, Amodei cited technical limitations rather than ethical objections. He pointed to "basic unpredictability" in current AI models as making it unsafe to remove humans from targeting decisions entirely. The company is "not categorically against fully autonomous weapons," he said, but believes the technology isn't reliable enough yet.

Amodei acknowledged the awkward position of a private company effectively setting military policy. "It's not tenable over the long term for a private company to decide these issues," he said, calling on Congress to establish guardrails for AI use in defense applications. He noted lawmakers have been slow to act even as the technology races ahead.

The CEO said he's unaware of any real-world cases where users have run up against Anthropic's restrictions, suggesting the dispute is more about future capabilities than current deployments.

The ban adds uncertainty for defense contractors already navigating AI integration. Companies will need to determine whether their Pentagon work involves Anthropic's technology—either directly or through subcontractors—and plan for alternatives during the six-month transition window. For CFOs, that means potential contract modifications, compliance reviews, and technology stack changes on government projects.