Examining Generative AI User Discontinuance from a Dual Perspective of Enablers and Inhibitors

As a negative behavior, user discontinuance may undermine user retention and reduce the competitive advantage of generative AI platforms. Extant research has focused on the positive behaviors such as user adoption and continuance of generative AI, and has seldom explored the formation mechanism underlying user discontinuance of generative AI. The purpose of this research is to explore generative AI user discontinuance intention from a dual perspective of enablers and inhibitors. We used a mixed method of structural equation modeling (SEM) and fuzzy-set qualitative comparative analysis (fsQCA) to conduct data analysis. The results revealed that discontinuance intention is influenced by both the enablers (misinformation, algorithm bias, low transparency, and dissatisfaction) and the inhibitors (perceived anthropomorphism, perceived intelligence, perceived interactivity, and flow experience). The results contribute to a comprehensive understanding of generative AI user discontinuance. They also suggest that generative AI platforms need to be concerned with both enablers and inhibitors in order to prevent user discontinuance and ensure a sustainable development.