How generative AI affordance drives users’ continuous usage intention: the mediation role of self-expansion and self-extension

The swift advancement of generative artificial intelligence (AI) is a key component of contemporary technological progress. Diverse generative AI tools are continually emerging and finding widespread application across numerous industries. Furthermore, individuals are increasingly exposed to and utilising these technologies. Nevertheless, the question of which technical attribute factors influence individuals’ intentions for continuous usage of generative AI remains unanswered in current research. Drawing from the affordance actualisation theory, this study focused on large language models (LLMs), a special type of generative AI tool, and formulated a theoretical framework delineating how affordances of generative AI impact users’ intentions for continuous usage, validated through the PLS-SEM method. The study reveals that the data capture, classification, delegation, and social affordances of generative AI have a positive impact on users’ self-expansion and self-extension. Self-expansion and self-extension in turn positively influenced users’ continuous usage intentions. Furthermore, data capture, classification, delegation, and social affordances exhibit significant indirect effects on users’ continuous usage intentions, mediated by two parallel factors: self-expansion and self-extension. This discovery contributes to ongoing research in the realm of emerging information technology, offering novel perspectives on how information technology affordances influence user responses.