Self-directed use of a generative artificial intelligence chatbot for mental health therapy and associated outcomes among university students in Tanzania

Background The use of Artificial Intelligence (AI) is on the rise across various aspects of people's lives, including in the assessment and management of mental health. Despite the growing incorporation of AI, there is still a limited understanding of how AI chatbots can be utilized by university students for self-directed mental health therapy. Method A cross-sectional study design was used, and was conducted among 1293 university students at the higher learning institutions in Dodoma, Tanzania. The sample size was calculated by G*Power, and the stratified sampling technique was used for the selection of universities. A structured self-administered questionnaire was used for data collection. The factors of AI chatbot usage and mental health outcomes were determined using multivariable logistic regression analysis. P < 0.05 was considered to be significant. Results Among 1294 study participants, 1267 (97.9%) were aware of AI chatbots. Regarding AI usage, 1221 (94.4%) reported using AI chatbots, while ChatGPT was the most commonly used chatbot by 1013 (78.3%) of the study participants. Students use AI chatbot as therapy for reducing stress, anxiety and depression. The Accessibility (AOR = 0.570, 95% CI = 0.329–0.987), Easy to use, Affordability (AOR = 0.294, 95% CI = 0.169–0.511), Privacy and anonymity (AOR = 0.315, 95% CI = 0.150–0.661), and Effectiveness of AI Chatbots (AOR = 0.361, 95% CI = 0.158–0.824) were the factors associated with AI Chatbot usage Conclusion Our findings demonstrated that AI chatbot is widely used by university students for mental health self-therapy, offering effective support in reducing stress, anxiety and depression symptoms. However, challenges like overreliance and privacy concerns persist. There is a need for the policies and curriculum developers to integrate AI use in various programmes, including mental health and psychiatric. Students should be taught the correct way of using AI in solving mental health-related issues