From Adoption to Knowledge Utilisation: Understanding the Role of Value, Usage, and Social Perception in Generative AI

Abstract

Generative AI (GAI) has fundamentally transformed how individuals acquire, generate, and apply knowledge in digital environments. While information systems research has extensively explained why individuals adopt new technologies, considerably less attention has been devoted to understanding how initial adoption develops into meaningful knowledge utilisation. Existing post-adoption research has primarily focused on continued system use; however, continued use does not necessarily indicate that AI-generated knowledge has become embedded in users' learning, decision-making, or problem-solving activities. This limitation highlights the need for a broader post-adoption perspective to explain how generative AI supports meaningful knowledge utilisation. To address this gap, this study develops a conceptual framework that extends post-adoption theory by explaining the mechanisms through which adoption intention evolves into knowledge utilisation intention. Drawing upon information systems, knowledge management, post-adoption theory, and social psychology, the proposed framework suggests that utilitarian value and hedonic value jointly influence adoption intention. Rather than assuming that adoption naturally leads to knowledge utilisation, this study argues that users gradually develop post-adoption perceived value through accumulated interaction with generative AI, which subsequently facilitates knowledge utilisation intention. Furthermore, usage frequency and pluralistic ignorance are introduced as behavioural and socio-cognitive boundary conditions that respectively strengthen and weaken this transformation. Together, these mechanisms explain why individuals with similar adoption intentions may demonstrate substantially different levels of knowledge utilisation. By distinguishing knowledge utilisation from continued technology use, repositioning perceived value as a post-adoption evaluative mechanism, and integrating behavioural reinforcement with social perception, the proposed framework offers a more comprehensive explanation of post-adoption behaviour in the context of generative AI. The framework contributes to both information systems and knowledge management research by providing a theoretical foundation for understanding how generative AI becomes meaningfully embedded in users' everyday knowledge practices and offers directions for future empirical research.