Canada’s Minister of Artificial Intelligence, Evan Solomon, says OpenAI has committed to implementing stronger safeguards for ChatGPT following a meeting with CEO Sam Altman over concerns about how the company handled warning signs linked to the Tumbler Ridge shooting.
The meeting took place on March 4 after Canadian officials questioned why OpenAI did not notify law enforcement earlier about an account associated with the shooter. The company had previously banned a ChatGPT account connected to 18-year-old Jesse Van Rootselaar in June 2025 but did not alert police because it determined there was no immediate or credible plan for real-world violence.
The issue gained attention after the Feb. 10 mass shooting in Tumbler Ridge, British Columbia, which left nine people dead and 27 others injured. Authorities later identified the banned account as belonging to Van Rootselaar. OpenAI said it contacted the Royal Canadian Mounted Police (RCMP) only after learning the shooter’s identity through media reports.
Following discussions with Solomon, OpenAI agreed to introduce several measures aimed at improving safety oversight in Canada. One of the key steps includes creating a direct point of contact with the RCMP to ensure faster communication when potentially dangerous activity is detected on the platform.
The company will also introduce safety protocols designed to guide individuals who appear to be in emotional distress toward local mental health and support services.
Solomon said he also requested that OpenAI apply its updated safety standards retroactively by reviewing previously flagged accounts. The goal is to determine whether any past cases would meet the company’s new criteria for reporting to law enforcement.
“If earlier incidents should have been referred under the new standards, they will be identified and promptly reported,” Solomon said in a statement.
OpenAI has also indicated it will explore involving Canadian experts in privacy protection, mental health, and law enforcement to help assess high-risk cases involving users in Canada.
Additionally, the company plans to provide a comprehensive report outlining new systems designed to detect repeat policy violators and potentially dangerous users. Solomon said the Canadian AI Safety Institute will review the company’s approach and provide technical recommendations to the government.
“These are immediate steps to improve safety and accountability,” Solomon said, adding that while artificial intelligence offers significant opportunities for Canada, strong safeguards and clear rules are necessary to maintain public trust.
In a letter sent to Solomon on Feb. 26, OpenAI vice-president for global policy Ann O’Leary acknowledged that the company’s approach would likely have been different under its updated policies.
“With what we’ve learned and our enhanced law-enforcement referral protocol, we would refer the account banned in June 2025 to authorities if the same situation occurred today,” O’Leary wrote.
The letter followed a Feb. 24 meeting in Ottawa where Solomon and other federal ministers, including Public Safety Minister Gary Anandasangaree and Culture Minister Marc Miller, asked OpenAI representatives to explain how the company evaluates threatening content and determines when to alert law enforcement.
After that meeting, Solomon expressed disappointment that the company had not presented new safety proposals immediately but said OpenAI promised to return with more detailed plans specific to Canada.
Meanwhile, British Columbia Premier David Eby has called for a nationwide reporting standard for AI companies, arguing that technology firms should not be allowed to set their own thresholds for notifying authorities.
The federal government has not yet confirmed whether it will introduce regulations for AI chatbots, but Solomon said officials are considering stronger privacy rules, new protections against online harms, and additional transparency requirements for AI platforms operating in the country.
Opposition Conservatives have called for an independent public inquiry into the tragedy, focusing on the shooter’s mental health history, firearm access, and prior interactions with law enforcement.
British Columbia’s chief coroner has also announced a formal inquest to examine the circumstances surrounding the shooting and identify possible systemic failures.
The attack in Tumbler Ridge claimed the lives of nine people, including a teacher and five students aged 12 to 13 at Tumbler Ridge Secondary School. Two family members of the shooter his mother and half-brother were also killed at their home. Police say Van Rootselaar later died from a self-inflicted gunshot wound.
Investigators say the teenager had a history of mental health struggles, and RCMP officers had previously visited the household multiple times due to concerns related to firearms and mental health issues.
Authorities hope the ongoing investigations and policy discussions will lead to recommendations aimed at preventing similar tragedies in the future.
