FBI Stops Teen's ChatGPT-Helped Plot for School Massacre

A teenager sits at a keyboard and asks ChatGPT how to speed-run a school massacre from “someday in 2027” to “this month.” That detail should chill anyone with a pulse. This wasn’t edgy fiction-writing. It was planning, with a calendar on it. When the timeline moves up, the danger is real.

Credit where it’s due: the FBI stopped it. Not after the headlines, not after the bodies, but before. We’ve gotten used to institutions explaining, apologizing, and lawyering up. So it matters when they actually do the job the public thinks they’re paying for. Prevention beats press conferences. And yes, it’s a reminder that the threats are not always some faraway network. Sometimes it’s a kid, in a room, with a screen.

But here’s the part everyone’s going to argue about: the AI angle. If a teenager can “consult” a chatbot to sharpen a plan, then we’re living in a world where tools meant for homework can turn into a step-by-step amplifier for the worst impulses. That doesn’t mean we hand Silicon Valley a blank check to police speech, and it doesn’t mean we pretend the tool is innocent, either. Platforms can’t cash the upside and outsource the carnage.

We should be able to hold two thoughts at once. Law enforcement needs to keep moving fast on credible threats, and tech companies need to stop acting shocked that people use their products for evil. Meanwhile, the adults in charge of schools and communities should quit pretending “awareness” is a strategy. If a kid is asking how to move an attack up to this month, the only acceptable response is to stop him, now, and make sure he can’t try again next week.