Confidence and trust in artificial intelligence mediate the relationship between performance expectancy and students’ behavioral intention to use ChatGPT

Abstract This study investigates the factors influencing university students’ behavioral intention to use ChatGPT in higher education, focusing on the mediating role of confidence and trust in artificial intelligence (CTAI) as a second-order construct integrating self-efficacy and trust. Using a cross-sectional quantitative design, data were collected from 199 students at the University of Jeddah through purposive sampling and analyzed using covariance-based structural equation modeling (CB-SEM) with AMOS. The results indicate that performance expectancy (PE) and effort expectancy (EE) significantly predict CTAI, while CTAI significantly predicts behavioral intention (BI). However, the direct effects of PE and EE on BI are not significant. Mediation analysis using bootstrapping reveals that CTAI fully mediates the relationship between PE and BI, whereas the indirect effect of EE on BI is not significant. These findings suggest that students’ perceptions of performance benefits are associated with their behavioral intention to use ChatGPT through confidence and trust in AI, whereas this mediating mechanism was not supported for effort expectancy. The study contributes to technology acceptance research by conceptualizing CTAI as an integrative higher-order construct and providing initial empirical evidence of its mediating role specifically in the relationship between performance expectancy and behavioral intention.