Pentagon Threatens 'Security Risk' Label to Force Anthropic AI Access

High-Stakes Standoff Between Pentagon and AI Developer

The U.S. Department of Defense has issued a severe ultimatum to artificial intelligence company Anthropic, threatening to designate the firm as a risk to the military's supply chain. This aggressive maneuver is intended to force the company to remove usage restrictions on its AI model, Claude. The deadline for compliance was set for Friday evening, marking a significant escalation in the tension between national security interests and AI safety advocacy.

Unprecedented Use of Designation and Cold War Laws

Defense Secretary Hegseth reportedly warned that failure to comply would result in Anthropic being labeled a supply chain risk—a designation historically reserved for foreign entities with ties to U.S. adversaries. Such a classification would effectively blacklist the company, preventing it from partnering with other organizations. Concurrently, the Pentagon signaled its readiness to invoke the Defense Production Act (DPA). This Cold War-era legislation grants the government authority to compel private companies to prioritize contracts deemed necessary for national defense.

Dispute Over Safety Safeguards

At the heart of the conflict are Anthropic's safety protocols. A company spokesperson revealed that the Pentagon’s proposed contract language failed to address critical concerns regarding the use of Claude for mass surveillance of American citizens or in the deployment of fully autonomous weapons. While the Defense Department framed the latest offer as a compromise, Anthropic argued that the legal wording allowed these essential safeguards to be bypassed at will.

Contradictions in National Security Strategy

Anthropic CEO Dario Amodei publicly highlighted the paradoxical nature of the government's threats. In a blog post, he pointed out the contradiction between labeling Anthropic a security risk while simultaneously declaring its technology essential for national defense. Legal experts and AI policymakers have noted this discrepancy, questioning the logic of maligning a company's reputation while relying on its assets for critical operations.

Despite the stringent deadline, negotiations remain active. Anthropic has expressed a strong preference to continue serving the military with safety measures intact, leaving the door open for a potential resolution before the deadline passes.