Determinants of Higher Education Students’ Use of Generative AI Chatbots: An Extended Technology Acceptance Model (TAM) Perspective

This study investigates the factors influencing higher education students’ self-reported use of generative AI (Gen-AI) chatbots through an extended Technology Acceptance Model (TAM). The model incorporates trust and individual impact alongside perceived usefulness and perceived ease of use to better explain students’ adoption behavior. Usage is defined as the self-reported frequency of chatbot use rather than post-adoption continuance intention. A quantitative, cross-sectional survey was conducted with 303 higher education students. Data were analyzed using Structural Equation Modeling (SEM) in Smart-PLS after confirming the reliability and validity of the measurement model. It's shown that perceived ease-of-use significantly affects both perceived usefulness and self-reported usage. Perceived usefulness also positively influences usage frequency. Trust shapes students’ perceptions of ease of use and usefulness but does not directly affect usage. Moreover, usage has a strong positive impact on individual outcomes, indicating academic and personal benefits associated with frequent use. Ease of use and perceived usefulness are the key drivers of students’ Gen-AI use. Trust influences adoption indirectly by shaping these perceptions. Sustained use of these tools enhances academic and personal outcomes, and the extended TAM proves to be a suitable framework for explaining Gen-AI adoption in higher education contexts.