Particle.news
Download on the App Store

AI ‘Unmasking’ Images After Minneapolis ICE Shooting Debunked as Grok Fabrications

Experts say generative edits invent facial details that mislead viewers, driving harassment.

Overview

  • Fact-checkers concluded the widely shared unmasked photos of the ICE agent were generated by xAI’s Grok from screenshots of masked video, and authentic footage shows the agents’ faces remained covered.
  • The false name “Steve Grove” spread with the AI image and triggered harassment of unrelated individuals, including a Missouri gun-shop owner and the publisher of the Minnesota Star Tribune.
  • MPR News and other outlets reported the agent was identified as Jonathan Ross, citing court records that also describe a prior 2025 traffic-stop incident in Bloomington.
  • AFP documented Grok’s role in producing multiple fabrications and reported receiving an automated reply from xAI stating, “Legacy Media Lies.”
  • Researchers, including Hany Farid, warned that AI ‘enhancement’ hallucinates facial details and noted dehumanizing manipulations of images of Renee Nicole Good and a woman wrongly identified as the victim.