Elon Musk's AI tool Grok was used in strikes against Iran, says U.S. government

3 days ago 2
ARTICLE AD BOX
A adjacent  state  of U.S. President Donald Trump, Musk folded xAI into his abstraction  exploration institution  SpaceX successful  February [File]

A adjacent state of U.S. President Donald Trump, Musk folded xAI into his abstraction exploration institution SpaceX successful February [File] | Photo Credit: REUTERS

Elon Musk's artificial quality instrumentality Grok was utilized successful strikes against Iran, the United States authorities revealed successful a ineligible briefing seen Tuesday by AFP.

The June 15 little defends the state turbines utilized by a elephantine information centre belonging to the trillionaire's institution xAI, which are the people of an biology lawsuit.

In the brief, the US Department of Justice argued that the suit "threatens American national, economic, and vigor information by seeking to unopen disconnected the powerfulness proviso for artificial quality innovation that supports the Department of War's subject operations."

To enactment the argument, national prosecutors presented grounds from Pentagon AI main Cameron Stanley successful which helium states, nether oath, that Grok is already successful usage wrong Project Maven, the U.S. military's AI-assisted targeting programme that was initially powered by Anthropic's Claude model.

The project's Maven Smart Systems (MSS) "enabled US forces to deploy implicit 2,000 munitions to 2,000 chiseled targets wrong 96 hours during Operation Epic Fury," Stanley's connection said.

Stanley praised Musk's exertion and "the greatly accrued operational ratio made imaginable by the Grok Gov Model."

The NAACP, a civilian rights organisation defending Black Americans' rights, is suing xAI and accusing it of operating dozens of turbines without permits successful usurpation of the Clean Air Act.

The rights radical says they pollute bulk Black neighborhoods, but xAI says the turbines are impermanent and mobile, and truthful not taxable to regulation.

At the extremity of February, the U.S. authorities terminated its contracts with Anthropic aft it refused to let its tools to beryllium utilized for afloat automated strikes oregon the wide surveillance of Americans.

The Pentagon past turned to Anthropic's competitors, similar Google, OpenAI and xAI, to proceed its pursuit of AI.

At Google, much than 600 employees demanded the institution not supply AI to the subject for classified operations. Others person raised wide concerns astir AI's threats.

The U.S. military's modulation to AI is taking time, and successful March the authorities had to admit that Claude was inactive being utilized for the warfare successful Iran.

A adjacent state of U.S. President Donald Trump, Musk folded xAI into his abstraction exploration institution SpaceX successful February, which carried retired the largest IPO successful past connected June 12.

Published - June 17, 2026 09:06 americium IST

Read Entire Article