Anthropic CEO stands firm as Pentagon deadline looms

Anthropic CEO Dario Amodei is holding his ground against the Pentagon, publicly stating he “cannot in good conscience” grant the military unrestricted access to the company’s AI systems. The announcement comes just ahead of a 5:01 p.m. Friday deadline set by Defense Secretary Pete Hegseth. This high-stakes standoff marks a critical moment for the AI industry, as a leading developer directly challenges its most powerful government client on the ethical boundaries of AI deployment in national security.

The core of the dispute lies in two specific use cases Anthropic has prohibited: the mass surveillance of Americans and the operation of fully autonomous weapons without a human in the loop. In response to Anthropic's refusal, the Department of Defense has threatened to either label the company a “supply chain risk” or invoke the Defense Production Act to compel cooperation. Amodei highlighted the contradiction in these threats, noting, “One labels us a security risk; the other labels Claude as essential to national security.” While Anthropic is currently the only frontier AI lab with systems ready for classified military work, the DOD is reportedly preparing rival firm xAI as a potential alternative.

This conflict illustrates the growing friction between the safety-focused culture of prominent AI labs and the operational demands of the national security apparatus. Amodei has signaled a willingness to end the relationship amicably, offering to “enable a smooth transition to another provider” if the Pentagon terminates their contract. The outcome will likely set a precedent for how AI companies navigate lucrative but ethically complex defense contracts. If the Pentagon follows through on its threats or pivots to a more compliant provider like xAI, it could reshape the relationship between Silicon Valley and the U.S. military for years to come.

This is a landmark test of corporate-state power dynamics in the AI era. Anthropic is betting its unique position as the sole classified-ready provider gives it enough leverage to enforce its ethical red lines, while the Pentagon is signaling it will not allow private companies to dictate the terms of national security. The outcome will define the rules of engagement for all AI labs seeking lucrative, but contentious, defense contracts.