Conversational AI chatbots in smoking cessation: a systematic review of behaviour change mechanisms and user experiences

Aim To synthesise evidence on experiences of artificial intelligence (AI) chatbot-supported smoking cessation interventions and evaluate the extent to which behaviour change theory is integrated into their design and implementation. Subject and methods A systematic review was conducted of studies published between 2016 and 2026 that examined AI chatbot-supported smoking cessation interventions. Eligible studies explored behaviour change processes and reported outcomes related to user experience, engagement, continuance intention, or smoking reduction. Study identification, screening, quality appraisal, and synthesis were undertaken in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines to ensure methodological transparency and rigour. Results The systematic review found that perceived anonymity, accessibility, usefulness, and quality of interaction were key factors shaping users’ experiences and ongoing engagement with chatbot interventions. Personalised support, timely feedback, and conversational communication were associated with greater user satisfaction and stronger cessation outcomes. Behaviour change techniques were embedded within some interventions; however, the application of behaviour change theories was often inconsistent and insufficiently reported. The evidence also highlighted the need for greater attention to diverse socio-economic and geographical contexts, given the influence of digital health literacy on intervention uptake and effectiveness. Conclusion AI chatbot interventions show promise as accessible and acceptable tools for supporting smoking cessation. Positive user experiences were associated with sustained engagement and more favourable cessation outcomes. However, stronger theoretical integration, more consistent evaluation approaches, and greater consideration of socio-economic and geographical diversity are needed to strengthen the evidence base and support the development of effective, user-centred digital cessation interventions.