A significant majority of Canadians believe AI chatbot companies like OpenAI should be legally required to report suspected imminent violence to authorities, even as privacy concerns loom large over enforcement mechanisms. The debate intensified after OpenAI chose not to report disturbing conversations that ChatGPT had with the individual responsible for a February mass shooting in Tumbler Ridge, British Columbia, months before the killings occurred .

What Triggered Canada's Push for AI Accountability?

Following the Tumbler Ridge shooting, Canada's federal government reconvened an expert advisory panel on online safety to determine whether artificial intelligence technologies and chatbots should be included in upcoming online harms legislation. The government is expected to table a revised online harms bill this year, which may include mandatory reporting requirements for AI companies that detect warning signs of violence in user conversations .

OpenAI's decision not to escalate the shooter's conversations with ChatGPT to law enforcement has become a focal point in this regulatory discussion. The case raises fundamental questions about corporate responsibility, the limits of AI monitoring, and whether companies should act as de facto gatekeepers of public safety .

How Should AI Companies Balance Safety and Privacy?

According to a subscriber survey conducted by The Logic, 77 percent of respondents support requiring chatbot companies to report suspected imminent violence to authorities. Many survey participants emphasized that such reporting could "at the very least identify people in need of psychological intervention." However, the same respondents expressed deep skepticism about trusting these companies with sensitive data .

The tension between safety and privacy emerged clearly in the survey responses. Less than one percent of subscribers said they "greatly" trust tech giants or social media companies to safeguard personal information, while the remainder indicated only "somewhat" or "not at all" trust. One respondent captured the paradox bluntly: "Are you kidding?" .

Privacy advocates raised specific concerns about the practical implementation of violence reporting requirements. Some respondents questioned whether there are sufficient policing and governmental resources to properly assess conversations flagged to authorities. One subscriber noted, "I do not believe the federal government has the technical capacity or resources to handle the potential onslaught of complaints and concerns" .

What Other Online Harms Are Canadians Prioritizing?

Beyond violence reporting, the survey revealed broad public support for addressing multiple categories of online harm. The findings show strong consensus on which issues should be tackled through legislation:

• Sexual Exploitation of Children: Over 90 percent of respondents agreed that sexual victimization of children and image-based sexual abuse content should be included in the upcoming bill.
• Self-Harm and Bullying: 82 percent of readers supported inclusion of self-harm or bullying content targeting minors in the legislation.
• Violent Extremism: 76 percent of respondents wanted violent extremism addressed through the online harms framework.
• Hate Speech: 62 percent of survey participants supported including hate speech in the regulatory scope.

The government is also considering a ban on social media for children aged 14 and under as part of the broader bill, following Australia's model of an under-16 restriction. Sixty-four percent of readers supported this age-based ban, citing the documented harms of social media to underage users and the benefits of removing social comparison pressures .

What Are the Practical Challenges in Enforcing AI Accountability?

While public support for regulation is strong, survey respondents acknowledged significant implementation hurdles. Many noted that enforcement represents the real challenge, not merely writing the rules. One subscriber observed, "We can't regulate all the risks away," while another pointed out that "history shows that kids will always find a way to access what is 'banned'" .

Many

Age-verification technology emerged as a particularly contentious enforcement mechanism. Many respondents who opposed a social media ban cited privacy concerns around age-verification systems as their primary reason. Others supported the ban itself but expressed reservations about the technology required to enforce it. Subscribers warned that giving tech companies access to sensitive biometric data would introduce new security vulnerabilities, noting that "hackers and bad actors are becoming increasingly sophisticated" and that a data breach would be "a matter of when, not if" .

Many

The paradox of trusting companies with enforcement became apparent in reader comments. One subscriber wrote, "Tech companies already don't protect our data and yet we provide it to them consistently. Now they will use that failure to prevent a social media ban, and 'privacy experts' will support them." This observation highlights the fundamental credibility gap between public expectations and corporate track records .

What Does This Mean for OpenAI and Other AI Companies?

The Canadian regulatory push directly implicates OpenAI and other AI chatbot providers in a new accountability framework. If the government includes mandatory violence reporting in its online harms bill, companies like OpenAI would face legal obligations to monitor conversations, flag concerning content, and report to authorities. This represents a significant shift from the current voluntary approach .

The stakes extend beyond Canada. Similar regulatory discussions are underway in other jurisdictions considering restrictive measures on social media and online content. OpenAI's handling of the Tumbler Ridge case has become a cautionary example of what happens when companies prioritize operational autonomy over potential public safety obligations.

As the federal government prepares to table its revised online harms bill this year, the balance between innovation, privacy, and public safety will likely define the regulatory landscape for AI companies operating in Canada and potentially influence approaches in other countries grappling with similar questions about corporate responsibility in the age of artificial intelligence.