On Wednesday, March 4, AI minister Evan Solomon met with OpenAI’s Sam Altman, looking for ways to “strengthen safeguards and accountability in Canada for ChatGPT,” says Solomon. “During that discussion, I asked OpenAI to take several actions, which Mr. Altman has agreed to do.”
These include establishing a direct point of contact with the RCMP and implementing safety protocols that direct individuals experiencing distress to appropriate local support services.
“I have also asked OpenAI to apply what they say is their new safety standards retroactively and review previously flagged cases which Mr Altman confirmed they are undertaking. This will determine whether additional incidents that would have been referred to law enforcement under OpenAI’s new safety standards were missed, and ensure they are promptly reported to the RCMP.”
He says OpenAI has also 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. OpenAI will also provide a full report outlining the new systems it is developing to identify high-risk offenders and repeat policy violators.
“I will ask the Canadian AI Safety Institute (CAISI) to examine OpenAI’s model and provide expert technical advice to my office,” says Solomon.
“These steps represent immediate actions to strengthen safety and accountability.”
He says the government needs to balance the dangers of AI with the opportunities. “Canadians must be confident that these technologies operate under clear rules, strong safeguards, and real accountability when risks emerge. Our government is continuing to examine a range of measures to strengthen protections — including stronger privacy frameworks, enhanced protections against online harms, and new transparency expectations for AI systems operating in Canada.”
He says the government will continue to work with experts, industry, and provinces and territories, “to ensure Canadians can benefit from artificial intelligence while remaining protected from harms.”
In a statement to Tumbler RidgeLines (seriously? When did we become statement-worthy?) a spokesperson for OpenAI says the company “remains committed to continuing this work with the Canadian government going forward.”
Trent is the publisher of Tumbler RidgeLines.
