AI: the Pentagon confirms that Musk’s Grok was used in attacks against Tehran
A tool derived from Grok, the ‘Grok Gov Model’, is already being used within Project Maven, the military programme for AI-assisted target identification and selection
Elon Musk’s artificial intelligence system, Grok, was used in attacks against Iran. This was revealed by the US government in a legal submission filed in defence of the gas turbines at a massive data centre operated by xAI, the company owned by the billionaire, which is the subject of an environmental lawsuit.
The US Department’s report
In a submission filed on 15 June and seen by Agence France-Presse, the Department of Justice argued that this case ‘threatens the national, economic and energy security’ of the United States, as it risks disrupting the power supply to artificial intelligence infrastructure now used by the armed forces.
Confirmation from the Pentagon
To support this argument, the Ministry presented the testimony of Cameron Stanley, head of AI at the Pentagon. Stanley stated under oath that a tool derived from Grok, the ‘Grok Gov Model’, is already being used within Project Maven, the military programme for AI-assisted target identification and selection, which was initially based on the Claude model developed by Anthropic.
