Service Quality in the Age of Generative AI: Quality Management Perspective on Student Satisfaction, Loyalty

ABSTRACT The rapid integration of Generative AI (GenAI) into higher education has outpaced the development of frameworks for assessing its service quality. This research addresses this gap by examining whether traditional quality management models can explain student experiences with tools like ChatGPT. Based on a survey of 347 students and Structural Equation Modelling (SEM), we validate a streamlined four-dimensional quality structure: Credibility, Usability, Privacy, and Efficiency. Theoretical results challenge the anthropomorphic view of AI; the exclusion of human-centric dimensions (such as empathy) indicates that students perceive GenAI as a transactional tool rather than a collaborative partner. Empirical findings reveal a nuanced mechanism: while Efficiency is the primary driver of satisfaction, Credibility functions as a "must-be" attribute—it does not increase satisfaction but is essential for driving loyalty. These findings provide higher education institutions with a specific framework to manage GenAI adoption, prioritizing functional utility and data security over forced relational features.