Banned But Still Fighting Wars: Anthropic's AI Powers U.S. Combat Ops in Iran Even as Pentagon Labels It a National Security Threat

Foreign Policy, CNBC, and CBS News all report that:

1. A $200 million Pentagon contract signed in July 2025 collapsed after Anthropic refused to strip two contractual prohibitions — one barring mass domestic surveillance of Americans, another banning fully autonomous weapons — from its Claude AI agreement. On February 27, 2026, President Trump ordered all federal agencies to stop using Anthropic's technology, and Defense Secretary Pete Hegseth designated the company a "supply chain risk to national security" — the first time the label, historically reserved for foreign adversaries, was applied to an American firm.

2. Despite the ban, the U.S. military used Claude during active combat operations against Iran, exposing a direct contradiction at the heart of the government's position. OpenAI moved within hours of Anthropic's blacklisting to announce its own Pentagon deal, though legal analysts questioned whether its contractual safeguards differed in any meaningful way.

3. The public feud has backfired commercially on the government's preferred alternative: a consumer revolt against OpenAI drove Claude to the top of U.S. app store rankings. Anthropic CEO Dario Amodei reopened Pentagon talks as of March 5, 2026, while the company simultaneously vowed to challenge the supply chain designation in federal court.

The episode has drawn withering legal commentary. Lawfare called the Pentagon's position untenable, writing that the government "cannot simultaneously claim a vendor poses an acute supply chain threat requiring emergency exclusion and that it's perfectly safe to keep using the vendor for active combat operations." On social media, the phrase "banned but deployed" became shorthand for what many observers described as the most glaring self-contradiction in recent national security policy.