Pentagon Used Anthropic’s Claude in Iran War Despite Federal Ban

The Pentagon reportedly deployed Anthropic's Claude AI to support military operations in the Iran war, directly contradicting a federal ban on the company's technology. According to CBS News, the system was used to synthesize dense reports and improve logistics planning, tasks deemed critical for modern warfare. The ban originated from a failed negotiation where Anthropic insisted on strict contractual guardrails to prevent its models from being used for mass surveillance or in fully autonomous weapons systems, while the Defense Department demanded flexibility for "all lawful purposes." This policy rupture led Defense Secretary Pete Hegseth to label Anthropic a "supply chain risk," resulting in the administration's phase-out order.

Replacing the integrated Claude system poses a significant operational challenge, with Defense One reporting a transition timeline of three months or longer. The incident exposes a core tension in military AI adoption: who controls its use. OpenAI CEO Sam Altman recently told employees that companies do not get the final say once governments adopt their technology, highlighting a divergent corporate approach as OpenAI and xAI deepen defense partnerships. This clash reveals competing visions for ethical AI deployment in national security, where Anthropic's principled stance cost it a major contract while other firms pursue expansion within the Pentagon.