Association between Generative AI self-efficacy and Generative AI acceptance: The mediating role of Generative AI trust and the moderating role of Generative AI risk perception

Generative AI has garnered global attention. This study aims to explore the association between generative AI self-efficacy and generative AI acceptance among university students and underlying mechanisms. A sample of 353 university students in China was measured using scales for generative AI self-efficacy, generative AI trust, generative AI risk perception, and generative AI acceptance. Structural equation model and Model 1 from SPSS PROCESS 4.0 were employed to test the research hypotheses. The results indicated that a positive association between generative AI self-efficacy and generative AI acceptance among university students (β = 0.49, P < .001). Assistance, comfort with generative AI, and technological skills are positively association with generative AI acceptance. Generative AI trust mediated the relationship between generative AI self-efficacy and generative AI acceptance, as well as the relationship between anthropomorphic interaction and generative AI acceptance. Generative AI trust also mediated the relationship between technological skills and generative AI acceptance. Moreover, Generative AI risk perception negatively moderated the relationship between generative AI trust and generative AI acceptance. This study provides a theoretical foundation for further research on generative AI.