This study investigates the integration of artificial intelligence (AI) into social research by employing a chatbot-based survey instrument to assess multidimensional poverty among urban Asnaf communities in Malaysia. The research focuses on the role of religiosity as a key behavioural determinant influencing the three core dimensions of the Multidimensional Poverty Index (MPI): health, education, and living standards. It also examines whether age acts as a moderating variable that alters the strength of these relationships. Drawing on a quantitative cross-sectional research design, data were collected from 392 registered Asnaf individuals residing in Malaysia’s Federal Territories of Kuala Lumpur, Putrajaya, and Labuan. A unique feature of the study was the use of an AI-powered chatbot deployed through the WhatsApp platform to administer the survey. This digital approach ensured respondent anonymity, improved data accuracy through real-time validation, and offered logistical advantages for reaching low-income, urban populations. The measurement model incorporated validated constructs for Islamic religiosity comprising belief (aqidah), ritual practice (ibadah), and ethical behaviour (akhlaq) alongside MPI indicators adapted to the Malaysian context. Data analysis was conducted using Partial Least Squares Structural Equation Modelling (PLS-SEM), which demonstrated that religiosity has a significant negative relationship with deprivation across all MPI dimensions. Additionally, the moderating effect of age was confirmed: older participants exhibited stronger religiosity–poverty reduction associations, particularly in health and living standards. These results offer both methodological and theoretical contributions. Methodologically, the study showcases how AI tools like chatbots can be effectively utilised in sensitive, community-based research. Theoretically and practically, it affirms that religiosity can function as a form of social capital, promoting well-being, education, and modest consumption patterns among marginalised Muslim populations. The study concludes by advocating for the integration of digital technology and culturally embedded constructs in the design of future poverty alleviation policies and welfare programmes.