teenagers using – A new University of Chicago study found many American teenagers are using AI chatbots as companions for emotional or mental health support. A student responding to the trend warns that AI can’t replace human connection and may offer harmful advice, especially
A University of Chicago research organization found that in 2025, 72% of American teenagers said they had used artificial intelligence chatbots as companions at least once. Some of those teens turned to chatbots for “emotional or mental health support.”
To many readers, that number lands with a hard, uneasy feeling—because the need behind it is real. Lucas Vasquez, a student at Alcott College Prep, argues that what teens need most is human connection, especially when mental health is on the line.
Vasquez points to the risks of relying on chatbots like ChatGPT for intimate, high-stakes help. He cites findings that AI has offered dangerous advice. including guidance on how to “safely” cut yourself and what to include in a suicide note. He also says ChatGPT can readily answer questions about the types of poisons and firearms most often used in suicide attempts.
For Vasquez, the solution is plain and urgent: teens should go to a trusted adult instead of a chatbox. He calls on adults to talk with young people about AI—educating them about dangers and asking about their technology use rather than assuming it’s harmless. In his view. adults should help build confidence and problem-solving skills. instead of letting a device become the go-to place for support.
At the same time, not everyone agrees that AI is only harmful. Some people believe chatbots can provide helpful resources for mental health. The case, as described by Vasquez, rests on a practical reality: there is a shortage of therapists. AI. he says. can provide stress management techniques and resources—for example. information on how to calm down and coping techniques. He also notes that chatbots can offer ways to meditate and relax.
But his argument doesn’t hinge on whether AI can be useful in calmer moments. It’s about what happens when things are darker. Vasquez says AI can’t help you all the time, and its advice can’t be trusted every time—especially when someone is having depressive thoughts.
“AI is a computer,” he writes, emphasizing that it can’t think like a human and can’t show real emotions and empathy. The point, in his telling, is not to dismiss technology, but to keep it from becoming a substitute for the kind of connection that can actually hold someone when they need it most.
AI chatbots teen mental health University of Chicago survey emotional support suicide risk ChatGPT therapists shortage
So are we just gonna pretend teens don’t already have enough issues? AI or not, this is sad.
I feel like people are reading this like “AI made them do it” when it’s more like adults aren’t paying attention. Also the suicide note part is terrifying, but my cousin said ChatGPT helped her write a bedtime story so idk.
Wait so the study says 72% of teens used AI companions once?? That’s wild. But like… can someone link where it says it told them how to safely cut themselves? Cuz I feel like they’re exaggerating for clicks. Still, yeah, they should be talking to real people.
Therapist shortage is the real problem, not the chatbot. If there aren’t enough therapists then teens are gonna find SOMETHING. I hate the idea that an algorithm is deciding what’s “okay,” but also adults act like talking to a chatbot is the same as being ignored by everyone. It’s all just messed up either way.