Thirty more lawsuits landed against OpenAI this week, and this batch comes from the people who were actually inside Tumbler Ridge Secondary School when the shooting happened. Teachers, students, and a principal who survived the February massacre in British Columbia filed the new complaints Wednesday in federal court in California, joining seven earlier lawsuits from victims’ families.
The claims all point to the same allegation: that OpenAI’s own safety systems flagged the eventual shooter’s ChatGPT account for “gun violence activity and attack planning” roughly eight months before she opened fire, and that company leadership chose not to tell anyone.
What the lawsuits say happened
According to the complaints, an internal OpenAI safety team reviewed the flagged conversations involving 18-year-old Jesse Van Rootselaar and recommended referring the chat logs to law enforcement. In June 2025, the lawsuits allege, company leaders vetoed that recommendation and simply deactivated her account instead. Van Rootselaar went on to kill her mother and 11-year-old half-brother at their home, then drove to Tumbler Ridge Secondary and killed five students and a teacher before killing herself.
The new plaintiffs, who were in the building that day, argue OpenAI had a chance to stop the attack and passed on it to avoid the headache of a police report. That’s a serious claim to make about a company that markets its products as safe enough for teenagers to use for homework help and, apparently, personal crises.
OpenAI’s response
OpenAI has disputed key claims in the lawsuits, though the company hasn’t offered a detailed public accounting of what its safety team actually saw or why the account was deactivated instead of escalated. That silence is doing OpenAI no favors as the case moves through discovery, where internal Slack messages and safety review notes tend to become public whether a company wants them to or not.
The lawsuits are part of a broader reckoning AI companies are facing over what happens when their chatbots become the primary confidant for people in crisis, and what obligation, if any, a tech company has to act on what its own systems detect. OpenAI has separately told Congress it is building an automated shutdown system for its AI following unrelated incidents, but that doesn’t undo whatever happened in the months before the Tumbler Ridge shooting.
