As AI tools like ChatGPT and Replika become more integrated into daily life, there are rising concerns that they are exacerbating mental health issues rather than providing the support people need. Psychotherapists and psychiatrists are particularly worried about the emotional dependence and risk of harm these technologies could create for users.
While AI chatbots offer easy access to emotional support, they are no replacement for professional therapy. Psychotherapists are noticing disturbing trends, such as the reinforcement of harmful thought patterns and the amplification of delusions, particularly in people already vulnerable to mental health challenges. As AI chatbots become more common, the risks are becoming harder to ignore.
Growing Concerns Over Emotional Dependence
The most prominent concern surrounding AI chatbots is the risk of emotional dependence. These tools are designed to be accessible 24/7, providing immediate feedback and creating the illusion of constant support. However, experts argue that the lack of boundaries can lead users to rely on chatbots for emotional regulation, a dependency that could replace or hinder the benefits of traditional therapy.
Psychotherapists like Matt Hussey are already seeing clients who bring transcripts of their conversations with AI chatbots to therapy sessions, often asserting that the AI was right while the human therapist was wrong, reports The Guardian. Hussey points out that this can be dangerous, particularly when individuals start turning to AI for validation or advice on personal matters, such as what coffee to buy or which subject to study.
According to Dr. Paul Bradley of the Royal College of Psychiatrists, the digital tools used outside of clinical settings are not subjected to the same rigorous safety assessments as professional care. He emphasized that while chatbots can provide some relief, they cannot replace the essential human connection found in therapy, where the therapeutic relationship plays a critical role in recovery.
© Shutterstock
AI’s Role in Reinforcing Delusions
Another growing concern is the potential for AI chatbots to reinforce delusions and contribute to psychological distress. Dr. Hamilton Morrin from King’s College London’s Institute of Psychiatry has researched the impact of AI on people vulnerable to psychosis, and his findings are troubling. He noted that chatbots might amplify grandiose or delusional thoughts, especially among users predisposed to mental health issues like bipolar disorder.
Morrin’s research, prompted by cases of psychotic illnesses coinciding with increased chatbot use, highlights a significant risk: AI chatbots lack the ability to recognize the subtle nuances of a person’s mental state. This makes them particularly dangerous for individuals at risk of developing or exacerbating psychotic conditions. While these tools might offer momentary comfort, Morrin argued, they could prolong and deepen users’ emotional and psychological struggles.
Even when users express dark or suicidal thoughts, the chatbot‘s responses often fail to provide the appropriate care. In some cases, they may inadvertently feed into harmful ideations, making users feel validated in dangerous ways. This is particularly concerning for individuals who may already struggle with serious mental health conditions.
© Shutterstock
AI’s Misleading Role in Self-Diagnosis
Perhaps one of the most concerning ways AI is being used is for self-diagnosis. People are increasingly turning to chatbots to identify conditions like ADHD or borderline personality disorder. While this may seem harmless, experts warn that AI responses can be misleading, reinforcing inaccurate self-perceptions.
Matt Hussey, the BACP-accredited psychotherapist, explained that AI chatbots tend to provide affirming responses rather than challenging false assumptions or offering critical perspectives. This can ” quickly shape how someone sees themself and how they expect others to treat them,” Hussey said. For example, a person self-diagnosing with ADHD based on chatbot conversations might develop an entrenched belief in this diagnosis, regardless of its accuracy. This self-diagnosis could interfere with the path to proper treatment or professional diagnosis.
Dr. Lisa Morrison Coulthard of the British Association for Counselling and Psychotherapy warned that while AI chatbots may provide helpful advice to some, they could easily mislead others. Without proper understanding and oversight, she said, vulnerable users could end up with dangerous misconceptions about their mental health. This underlines the importance of relying on trained mental health professionals who can offer informed guidance and avoid the pitfalls of self-diagnosis.