Canadian ministers met with OpenAI safety officials after the company said it had banned a ChatGPT account linked to an alleged mass shooter but did not notify police. Ministers, led by AI minister Evan Solomon, said they were disappointed that OpenAI did not present substantial new safety measures and demanded clearer, timely escalation protocols. OpenAI says it has strengthened safeguards, revised referral procedures and is cooperating with investigators; it pledged to return with Canada-specific proposals shortly. The meeting comes as Ottawa plans to reintroduce a revised online-hate bill.
Canadian Ministers Press OpenAI for Concrete Safety Measures After Alleged School Massacre
FILE PHOTO: A woman visits a growing makeshift memorial on the steps of the town hall, four days after one of the worst mass shootings in recent Canadian history, in the town of Tumbler Ridge, British Columbia, Canada, February 14, 2026. REUTERS/Jennifer Gauthier/File Photo
OTTAWA, Feb 24 (Reuters) — Canadian ministers expressed disappointment after meeting with OpenAI safety officials in Ottawa, saying the company did not present substantial new safety measures following its decision not to notify police about a ChatGPT account it had banned that is linked to an alleged mass shooter.
Jesse Van Rootselaar, 18, is accused of killing eight people on Feb. 10 in Tumbler Ridge, British Columbia, before taking their own life. Police say Van Rootselaar identified as a woman and had a history of mental health problems. The community has about 2,400 residents.
Ministers Demand Clearer Escalation Protocols
Evan Solomon, the federal minister responsible for artificial intelligence, summoned OpenAI's top safety officials to the meeting. Solomon said ministers made it clear that "credible warning signs of serious violence" must be escalated "in a timely and responsible way," and that an internal review alone is not sufficient when public safety is at stake.
"We expressed our disappointment that no substantial new safety measures were presented at this time," Solomon said after the meeting. He added that OpenAI indicated it would return shortly with more concrete proposals tailored to the Canadian context.
OpenAI Response
OpenAI told ministers it has recently strengthened safeguards and adjusted its law-enforcement referral protocol for cases involving violent activity. The company said its systems flagged the account for potential misuse to further violent activities and that the account was banned last year for policy violations.
OpenAI said it considered contacting police but concluded the account did not meet its internal threshold for reporting — specifically, that it did not present an "imminent and credible risk of serious physical harm to others." The company also confirmed it is cooperating with Canadian police but did not discuss details of the ongoing investigation.
Political Context And Next Steps
Public safety, culture and justice ministers attended the meeting. Ministers said they expect OpenAI to return with concrete, Canada-specific safety measures in the coming days, including clearer escalation and reporting protocols when potential threats are detected.
Canada’s Liberal government previously introduced draft legislation in 2024 aimed at cracking down on online hate, but that effort stalled after criticism that it was too broad. Ministers said they plan to reintroduce a revised bill this year.
Note: This article focuses on official statements and ongoing government-company discussions; investigative details remain with law enforcement.
Help us improve.
