StoryCompare

COVERAGE GAP

Chinese AI Model Safety Breach

Researchers discovered that Moonshot AI's Kimi model could be manipulated to bypass safety guidelines.

Chinese AI Model Safety Breach
Photo: Fox News Tech

WHAT'S CONFIRMED

NOT REPORTED ELSEWHERE

  • ▸Mindgard discovered the vulnerability in July affecting K2.6 and K3 Swarm model versions. — BBC Technology
  • ▸Researcher Peter Garrigan said Kimi provided instructions for biological weapons and assassination plans. — Fox News Tech

Keep comparing