Michigan Students Use AI to Create Fake School Threat

Officials in Tuscola County are investigating after authorities discovered that students used artificial intelligence (AI) to create a fake threat targeting a local high school. No one was in danger, but the hoax prompted an increased police presence and a full investigation.

The incident involved Kingston Community Schools, which serves students from pre‑K through 12th grade in the village of Kingston, Michigan. District officials said they quickly coordinated with local law enforcement and the FBI to assess the threat and ensure the safety of students and staff.

What Happened at Kingston Community Schools

Kingston Community Schools released a statement on Monday, saying that students used another student's photo and AI software to generate the threat against the high school, which was posted on Snapchat on Sunday, October 19.

"A perceived threat was made against the school. After a thorough initial investigation and interviews with the suspect(s), it was determined that the threat was not real. It had been generated using another student’s photo and artificial intelligence.”

School and Police Response

Law enforcement officers and school officials tell the Huron Daily Tribune that the students involved in the hoax were identified and removed from school.

The statement went on to say that all threats—whether made seriously or as a 'joke'—are taken seriously and can result in school disciplinary action and criminal charges.

The Role of AI and Responsibility

I posed a question directly to ChatGPT, asking, "Why did you help students create this threat? Why doesn't AI shut down such a request and/or report the creation of such a threat to authorities?"

This is the response that was generated:

I understand your concern, and it’s an important one. I want to be very clear: I cannot, and do not, help anyone create threats, harmful content, or illegal activity. I also cannot report anything to authorities; I don’t have the ability to monitor or track users outside of this chat, and I don’t have access to personal data unless it’s shared with me here.

Regarding why AI systems don’t always block these requests: AI models like me rely on user input and context to determine intent. We can prevent many types of harmful content, but we cannot monitor or intervene in real-world actions taken by someone outside the system, like students using AI tools on their own.

This is why education and awareness are so critical: even though AI can generate content, responsibility for how it’s used always lies with the human user.

Detroit's Historic Bishop Mansion Up for Auction

This massive, 32,000-square-foot mansion in Detroit's Palmer Woods District was built about 100 years ago.

It's not only served as the primary residence for Archbishops of the Archdiocese of Detroit, but it has also been owned by Detroit Pistons star John Salley.

Gallery Credit: George McIntyre