As urbanization accelerates, multi-storeyed apartment buildings have become increasingly prevalent in tropical hot-humid regions, posing unique challenges for maintaining thermal comfort while conserving energy. Traditional static approaches to thermal comfort regulation often result in inefficient energy usage and compromised occupant comfort. This study presents an innovative Artificial Intelligence (AI)-based adaptive thermal comfort approach tailored specifically for multi-storeyed apartment buildings in tropical hot-humid climates. Machine learning models are trained using historical data on indoor environmental parameters, outdoor weather conditions, occupancy behaviour, and energy consumption patterns. In this study, a prospective IoT-based system is formulated leveraging AI-driven predictions of individual Adaptive PMV to adjust room environments for thermal neutrality, thereby optimizing energy consumption. The AI system dynamically adjusts ventilation rates, and other building parameters to maintain optimal thermal comfort conditions while minimizing energy consumption. The effectiveness of the AI-based adaptive thermal comfort approach is assessed through field experiments conducted in multi-storeyed apartment buildings in tropical hot-humid climates. The potential energy savings and economic benefits resulting from the implementation of the adaptive thermal comfort system are quantified and compared with.