Pentagon Grok claim: Musk’s AI was used in strikes on Iran

A striking claim has emerged in the United States over the use of artificial intelligence in military operations. Cameron Stanley, director of the Pentagon’s Digital and Artificial Intelligence Office, said in a sworn statement submitted to court that Grok, developed by Elon Musk’s company xAI, was actively used in US national security and military operations. Stanley said Grok had been integrated into the Maven Smart System, which was used in attacks on Iran.

Grok Gov Model detail

Stanley said the Pentagon made use of the “Grok Gov Model” developed by xAI. He stated that the model had been shaped according to the needs of the US government and claimed that some of its features were not available in other artificial intelligence models.

Integrated into Maven system

According to the statement, the Grok model was used within the Maven Smart System, an AI-powered command and data analysis platform developed by the Pentagon and Palantir. Stanley said the system played a role in target detection, intelligence activities, military planning and logistics planning.

Claim of 2,000 targets in 96 hours

Stanley cited “Operation Epic Fury” as an example of Grok’s use. The statement said the US military, through the Maven platform, targeted 2,000 separate points with more than 2,000 munitions within 96 hours.

Claude model had also been mentioned

Earlier reports in the US press had claimed that the Pentagon used Anthropic’s Claude model and the Maven Smart System while planning strikes on Iran. Those claims suggested that Maven used advanced AI technologies to identify and locate potential targets in Iran.