Jesse Van Rootselaar, the suspect in the Tumbler Ridge, British Columbia mass shooting, was raising alarm among OpenAI employees months before the shooting happened. This June, Jesse had conversations with ChatGPT involving descriptions of gun violence that triggered the chatbot’s automated review system. Several employees expressed concern that her posts could be a harbinger of real-world violence and urged company officials to contact authorities, but they ultimately declined.
According to The Wall Street Journalcorporate executives at the company decided that Rootselaar’s posts did not pose a “credible and imminent risk of serious physical harm to others.” The company banned Rootselaar’s account, but it doesn’t appear to have taken any further action. We’ve reached out to OpenAI to ask who specifically made this decision and how it was made, and we’ll update if we hear back.
The decision not to alert law enforcement appears to have been misguided in retrospect, as on February 10, nine people were killed and 27 injured, including Rootselaar, in Canada’s deadliest mass shooting since 2020. Rootselaar was found dead at the site of Tumbler Ridge High School with an apparent self-inflicted gunshot wound, where most of the incidents occurred.