A Review of the Impact of AI Chatbots on Mental Health and Therapy

Mental health issues have been rising in prevalence among people today, being more widespread in teens, which can lead to social isolation or even death from suicide. People are turning to artificial intelligence (AI) chatbots as tools to ask for advice. The use of AI chatbots can result in the amplification of the feelings people expressed. Based on a literature review on the relationship between AI chatbots and mental health therapy, this study explores the increasing use of AI chatbots as a substitute for therapists and the potential effects of using AI for advice. Despite the increasing awareness about the dangers of the use of AI chatbots as therapists, there have been few precautions put in place. The most promising recommendation to prevent the use of AI chatbots as a therapist involves programming general-purpose generative and similar AI chatbots to completely shut down if they sense any mental health issues that risk the wellbeing of the user. Right now, AI chatbots only give cautionary warnings to the user, but when faced with a difficult situation, the user will often ignore these messages and continue asking for advice. Programming the AI to completely shut down limits peoples’ ability to rely on the AI as a therapist. Additionally, an alternative to people directly using AI for mental health guidance is the use of general-purpose generative AI chatbots as a tool for therapists via increased opportunity for training, providing a way to incorporate AI in the field without causing harm.