The US military reportedly relied on Anthropic’s Claude AI for intelligence analysis and targeting during an Iran strike hours after Trump ordered a ban on the company’s systems.