The US military reportedly used Anthropic's Claude AI for intelligence and targeting in Iran strikes, despite Trump's ban hours earlier. He called the company "Radical Left." The incident highlights challenges in rapidly removing embedded AI tools from military operations.