By Paul Vieira
OTTAWA--Canada says OpenAI Chief Executive Sam Altman has agreed to take immediate steps to strengthen safety protocols regarding notifying police about potentially suspicious use of the company's ChatGPT chatbot.
Artificial Intelligence Minister Evan Solomon added that he also asked Altman to apply these changes retroactively and to review previous incidents that may have been referred to law enforcement for further investigation.
Solomon's office issued a statement late Wednesday night summarizing a virtual meeting earlier in the day between the minister and the OpenAI CEO. The government is seeking changes to how OpenAI and other digital platforms operate following The Wall Street Journal reporting that indicated company employees raised alarm bells about interactions with ChatGPT in 2025 involving an individual whom police identified last month as a suspect in a fatal school shooting in Tumbler Ridge, British Columbia.
In 2025, OpenAI shut down Jesse Van Rootselaar's account, but it didn't notify the police. The company last week pledged to modify its protocols on alerting police, and acknowledged that under the changes the company would have alerted police about Van Rootselaar's interactions.
An OpenAI spokesperson said Altman advised Solomon that the company is recrafting its protocols under which interactions are referred to law-enforcement agencies, and improving systems to account for country and community context. "We remain committed to continuing this work with the Canadian government going forward," the company said.
The Tumbler Ridge shooting is drawing closer scrutiny about potential real-world harms from AI, analysts say, adding that a bold regulatory response from Canadian officials is required. The response "cannot simply be to require companies to monitor and report private conversations to law enforcement," said Taylor Owen, public-policy professor at Montreal's McGill University and founding director of the school's center for media and technology. "What is needed is a broader regulatory framework that addresses the upstream design decisions and safety architectures that allowed these situations to arise in the first place."
Michael Geist, an internet-law expert and law professor at the University of Ottawa, said any new regulatory measures must not violate individuals' privacy rights. He said any new regulatory measures should focus on "on ensuring there is full disclosure of user safety policies and how they are implemented and enforced."
Solomon has said his government is exploring options to ensure that digital platforms and emerging technologies, like AI, adopt standardized public-safety protocols.
As for his conversation with Altman, Solomon said Altman agreed to take several actions. For instance, Solomon said that OpenAI committed to assess "how they would include Canadian privacy, mental health and law enforcement experts into the process to identify and review high-risk cases involving Canadian users."
Solomon said OpenAI also pledged to provide a report outlining the new protocols it is developing to identify high-risk offenders and repeat policy violators.
OpenAI has agreed to cooperate with investigators and with authorities in British Columbia who are set to lead a public inquest into the shooting at the remote town of Tumbler Ridge, which left eight people dead and dozens injured. Police found Van Rootselaar dead at the scene.
Write to Paul Vieira at paul.vieira@wsj.com
(END) Dow Jones Newswires
03-05-26 1053ET
