Pentagon AI Chief: Grok Helped Bomb Thousands of Targets

The Pentagon is leaning on Elon Musk's artificial intelligence to make its case in court—and revealing a lot in the process. In a sworn filing, the Defense Department's top AI official said xAI's Grok chatbot was used to help fire more than 2,000 munitions at 2,000 targets in Iran over four days, and called the system's continued operation "a matter of paramount national security," the Independent reports.

• Grok, via a special "Grok Gov" model, is one of a handful of AI tools the Pentagon says can support sensitive and "mission-critical" work, including in top-secret environments. In the filing, Cameron Stanley said the strikes were "a testament to the greatly increased operational efficiency made possible by the Grok Gov Model," per the Hill.

The disclosure comes as the administration asks a federal judge to dismiss an NAACP lawsuit claiming xAI's data centers are running dozens of unpermitted gas turbines that pollute nearby Black communities in Mississippi and Tennessee. The Pentagon argues those facilities are vital national-security infrastructure, able to provide emergency power and preserve a technological edge.

Lawmakers, meanwhile, are pushing to rein in battlefield AI after strikes in Iran—including one on a girls' school that killed at least 175 people, most of them children—raised questions about whether automated targeting tools and outdated maps contributed. A bill backed by Democratic Sen. Kirsten Gillibrand would bar AI from controlling nuclear weapons, domestic surveillance, and fully autonomous weapons systems. "Right now, the Pentagon is moving toward deploying incredibly powerful AI technology without commonsense guardrails in place, which could have catastrophic consequences that make all of us less safe," she said earlier this month.

• Before Grok's involvement in strikes was revealed, some analysts suspected Anthropic's Claude chatbot, also used by the military, could be to blame for the strike on the Minab girls' school, Futurism reports. Other analysts suspected human error, noting that AI might have been better at determining it was not a military site. Asked about the strike Wednesday, Trump said an investigation is underway and the school was not hit "on purpose," the New York Times reports. "Mistakes are made. War is nasty," he said.