Conversational AI for HIV/AIDS education among youth: a systematic review of chatbots, mHealth tools, and digital interventions.

Young people aged 15-29 account for a substantial proportion of new HIV infections globally, yet traditional prevention approaches have struggled to achieve consistent reach. This systematic review synthesizes evidence on conversational AI, chatbots, mHealth apps, SMS-based tools, and web-based digital interventions for HIV education, prevention, or care among youth aged 10-29 years. A search of six databases (2016-2025) identified 14 eligible studies, assessed using MMAT and Cochrane RoB 2.0. Studies were conducted predominantly in the USA (n = 6) and Sub-Saharan Africa (n = 6). Tools included AI chatbots (n = 4), mHealth apps (n = 3), SMS interventions (n = 3), web-based platforms (n = 2), and multi-component interventions (n = 2). Chatbots demonstrated high feasibility, with 79-95% HIV self-test completion in descriptive studies. One pilot RCT found SMS interventions increased condom-protected sex (1.8x) and HIV testing (2.4x), requiring confirmation in larger trials. mHealth apps achieved 99.4% ART adherence message fidelity, a process measure only. Facilitators included privacy, anonymity, and youth-centered design; barriers included connectivity issues and unnatural conversation. Conversational AI and mHealth tools are feasible and acceptable for youth HIV prevention, but effectiveness evidence remains limited, with only one large RCT demonstrating impact on a biomedical outcome (STI incidence). Future research should prioritize rigorous effectiveness trials and equitable access strategies.