Particle.news
Download on the App Store

Report: OpenAI Declined to Alert Police After ChatGPT Flagged Suspect’s Violent Plans Months Before Canada Mass Shooting

OpenAI cites internal rules requiring a clear, imminent threat before notifying law enforcement.

Overview

  • Investigative reports say Jesse Van Rootselaar described detailed violent scenarios to ChatGPT around June, triggering automated flags.
  • Roughly a dozen OpenAI reviewers examined the chats, and several urged supervisors to contact Canadian police, according to the reporting.
  • Company leadership suspended the account but did not notify authorities, concluding the exchanges did not meet its credible and imminent risk threshold.
  • OpenAI reached out to the Royal Canadian Mounted Police only after the Feb. 10 Tumbler Ridge attack that left eight dead and at least 25 injured, with the 18-year-old suspect dead by suicide.
  • Separate records and posts show prior police welfare checks with temporary weapon seizures and online activity including 3D-printed ammunition claims, shooting-range photos, and a Roblox mall-shooting simulator.