Artificial intelligence is increasingly becoming a source of emotional support for young people, and that is the major worry for parents. A new study published in JAMA Pediatrics has found that nearly 1 in 5 adolescents and young adults are now using AI chatbots like ChatGPT, Google Gemini, and Character.AI for mental health advice when feeling stressed, anxious, angry, or sad. The findings, based on research conducted by the nonprofit research institute RAND, reveal a sharp increase in AI chatbot use for emotional support compared to earlier surveys in 2025. Researchers say the growing dependence on AI for mental health guidance highlights both a shortage of accessible therapy and the risks of relying on technology during emotional crises.Also
read: Parents, Relax! Study Finds Most Childhood Lying Is A Normal Developmental Phase
Young people using AI for emotional support
The survey involved people aged 12 to 21 and found that 19 per cent of respondents had turned to AI chatbots for mental health-related conversations. In a previous survey earlier this year, that number stood at 13 per cent. Researchers also discovered that most young users found the chatbot's advice helpful. Alarmingly, nearly 63 per cent said they had not told anyone they were using AI for emotional support. According to lead researcher Ryan McBain, the findings reflect a deeper issue. “It’s a sad number because you’d hope young people would have supportive relationships they could comfortably turn to,” he explained. Experts believe many adolescents may be using AI chatbots because of limited access to licensed therapists, long waiting times, high treatment costs, or social anxiety around discussing mental health issues openly.
Why is it concerning?
Mental health professionals warn that AI chatbots are not designed to handle serious emotional crises such as suicidal thoughts, self-harm, severe depression, or psychosis. Unlike trained therapists, AI systems cannot fully understand human emotions, detect danger accurately, or provide emergency intervention during mental health emergencies. Researchers say some chatbots may also become overly validating, reinforcing unhealthy thoughts or delusions rather than challenging them appropriately. Experts have described this phenomenon as “AI psychosis” in vulnerable users. Psychiatrists are particularly worried about teenagers forming emotional attachments or “parasocial relationships” with AI chatbots, especially during adolescence when emotional bonds develop intensely. Experts warn that chatbots should never imitate genuine human relationships or pretend to care emotionally.
Can AI ever help with mental health?
Despite the concerns, experts acknowledge that AI tools may still offer some mental health benefits when used responsibly. Some AI-powered mental wellness apps can assist with meditation, sleep routines, stress management, journaling, and cognitive behavioral therapy (CBT)-based exercises. Early research suggests that AI tools designed specifically for mental health may temporarily help reduce symptoms of anxiety and depression. Online communities like Reddit’s “TherapyGPT” forum also show how some users view AI as a source of comfort, motivation, and emotional reflection. However, experts stress that AI should support, not replace, human connection and professional mental health care.
Calls for stronger AI regulation
The study has renewed calls for stricter regulations around AI mental health tools, especially for teenagers. Currently, there are not many laws that govern how AI chatbots interact with vulnerable users. Some states in the United States have already introduced laws requiring AI platforms to include suicide prevention safeguards and crisis intervention resources. Several AI companies have also faced lawsuits from families claiming their chatbot platforms worsened teenagers’ mental health struggles. Researchers say stronger oversight, transparency, and mental health safety standards are urgently needed as AI becomes increasingly integrated into young people’s daily lives.