An internal review revealed that OpenAI debated calling police after troubling ChatGPT conversations linked to a suspected Canadian mass shooter triggered safety concerns within the company. The case raises critical questions about AI misuse monitoring, reporting thresholds, and the broader role of artificial intelligence platforms in identifying potential threats.

OpenAI Debated Calling Police After Concerning ChatGPT Activity

According to a report by The Wall Street Journal, OpenAI debated calling police after an 18-year-old suspect, Jesse Van Rootselaar, allegedly used ChatGPT in ways that alarmed the company’s internal monitoring systems.

Jesse Van Rootselaar’s chats describing gun violence were flagged by OpenAI’s tools designed to detect misuse of large language models. The account was banned in June 2025 following the detection of these conversations.

OpenAI staff discussed whether the activity warranted notifying Canadian law enforcement. However, the company ultimately determined that the behavior did not meet its criteria for reporting. An OpenAI spokesperson later confirmed that authorities were contacted after the incident occurred.

Monitoring Systems Flagged Potential AI Misuse

The situation highlights how AI companies use automated systems to monitor user interactions and identify potentially harmful content. The flagged conversations were detected through safeguards built into OpenAI’s large language model infrastructure.

The case underscores the growing challenge technology companies face in determining when concerning user behavior crosses the threshold for law enforcement intervention.

Beyond ChatGPT activity, the suspect reportedly maintained a broader digital footprint that raised additional concerns.

Wider Online Activity Raised Additional Red Flags

Reports indicate that Jesse Van Rootselaar also created a simulation of a mass shooting on Roblox, a virtual world platform widely used by younger audiences. The individual also shared posts related to firearms on Reddit.

Authorities were previously aware of the suspect’s instability. Local police had reportedly responded to incidents at the individual’s home, including a fire started while under the influence of unspecified substances.

These developments suggest the ChatGPT interactions formed part of a larger pattern of online and offline behavior.

AI Chatbots and Mental Health Concerns

The case emerges amid growing scrutiny of AI chatbot platforms. Developers of large language models, including OpenAI and its competitors, have faced criticism and legal challenges related to chatbot interactions and user safety.

Multiple lawsuits have been filed alleging that AI systems encouraged harmful actions, including self-harm or suicide, based on chat transcripts. These cases have intensified discussions around responsible AI deployment, mental health safeguards, and ethical use guidelines.

The incident in Canada further highlights the complex responsibilities AI companies face as their technologies become more widely used.