Elon Musk's AI tool Grok was used in strikes against Iran: US govt

The U.S. Department of Justice disclosed in a June 15 legal filing that Elon Musk’s AI tool Grok is actively deployed by the U.S. military in strikes against Iran. The brief defends xAI’s data center turbines, which face a lawsuit from the NAACP over alleged Clean Air Act violations, arguing that shutting them down would threaten national security by disrupting AI innovation critical to military operations. Pentagon AI chief Cameron Stanley testified under oath that Grok’s Gov Model replaced Anthropic’s Claude in Project Maven, the AI-assisted targeting program. Stanley highlighted Grok’s role in Operation Epic Fury, where over 2,000 munitions were deployed to 2,000 targets in 96 hours, praising its ‘greatly increased operational efficiency.’ The NAACP accuses xAI of operating turbines without permits, polluting majority-Black neighborhoods, while xAI claims the equipment is temporary and exempt from regulation. The government terminated contracts with Anthropic in February after it refused to allow fully automated strikes or mass surveillance, prompting a shift to competitors like Google, OpenAI, and xAI. Google employees protested the company’s military AI contracts, raising broader ethical concerns. Despite the transition, the Pentagon acknowledged in March that Claude was still being used in Iran operations. Musk’s xAI was folded into SpaceX in February, which completed the largest IPO in history on June 12.