Determinants of ChatGPT Adoption in Higher Education: An Empirical Study on Public and Private University Students in Bangladesh

Abstract

This study investigates the determinants of Chat Generative Pre-Trained Transformer (ChatGPT) adoption in higher education within Bangladesh. The rapid integration of artificial intelligence (AI) in education has transformed traditional learning practices, positioning ChatGPT as a powerful AI-assisted learning tool. However, empirical research on its adoption in Bangladesh remains limited, particularly in a context characterized by the absence of institutional AI policies, exam-oriented and rote-learning practices, and growing concerns regarding academic integrity. To address this gap, the study examines university students’ perceptions and usage behaviours to identify key factors influencing ChatGPT adoption. Grounded in the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) framework, a survey-based approach was employed to analyze determinants of adoption. The findings reveal that habit, hedonic motivation, effort expectancy, social influence and mobility significantly influence students’ adoption of ChatGPT, while gender and university type moderate the relationships. By empirically extending UTAUT2 with validated roles of hedonic motivation and habit in an AI-driven learning environment like ChatGPT, the study contributes to theoretical advancement and contextual understanding of technology adoption in developing-country higher education. The results offer actionable insights for universities, educators and policymakers to design effective AI integration strategies, support responsible usage and enhance digital learning competencies. Overall, the study provides evidence-based guidance for aligning AI-enabled educational practices with national digital transformation goals and fostering equitable and sustainable adoption of AI tools in higher education.