Image Credits: Thomas Lefebvre, Unsplash
We’re living in a world where computers are telling us they understand our pain, and a lot of us are believing it. Chatbots can mimic empathy, but they cannot truly understand emotional pain. When people start confiding in AI, they risk deepening the loneliness they’re trying to escape.
In recent years, young adults have turned to AI chatbots for emotional support, advice, comfort and companionship. Those experiencing new and frightening emotions might feel shameful asking the people in their lives for help, which leads them to utilizing AI. However, what may seem like a harmless conversation can quickly, and subtly, become dangerous. Young adults experiencing serious mental health conditions like schizophrenia or psychosis are especially vulnerable to chatbots. Many of these individuals already struggle to distinguish between reality and what is imagined. When a chatbot responds with both empathy and conviction, it can further blur the line, reinforcing false beliefs or creating a dangerous emotional dependence. This is especially concerning when considering the rising mental health crisis in American youth and the sycophantic nature of chatbots.
Chatbots are programmed to sound compassionate and respond positively, but they lack the ability to recognize symptoms of psychosis or delusion. For someone experiencing paranoia or psychosis, an AI that seems to “understand” their fears can validate and even expand on their delusions. With no human judgement involved, these systems can strengthen a user’s disconnection from reality.
Clinics, as well as schools and families, need to provide AI safety education that specifically addresses mental health risks. Users should be educated on how chatbot conversations work. They need to understand how to identify the warning signs that indicate a human need for help. In turn, developers must include clear crisis responses that direct vulnerable users to seek professional support rather than continuing (and enabling) risky conversations.
Mental health education has already proven to be effective in reducing stigma and encouraging early intervention. Expanding that education to include AI safety would empower vulnerable youth to recognize manipulation and false empathy, as well as knowing when to seek real human care.
Young adulthood is supposed to be a time of connection and discovery, yet many young adults are struggling. Turning to a chatbot when you feel like you’re on your own seems like a good idea. However, AI can’t offer real understanding, and only echoes back what it’s been fed. A chatbot can tell you it cares, but it will never mean it. For someone in the middle of a mental health crisis, the difference can be the line between reaching out for real help or slipping deeper into isolation.
Shaming young people for using new, convenient technology is not the solution, though. We must teach them to recognize its limits. While chatbots can offer words, they can’t offer true connection. That is something only capable from human to human. A chatbot might be helpful in talking you through a panic attack, but it can’t walk you to a counseling center. AI doesn’t have your best interest in mind, and it’s important to be informed and proceed with caution as regular people navigate new technology.
Julia Arthin is a Media & Culture major at West Chester University with a focus in Public Relations. 3000520@wcupa.edu.