Generative Artificial Intelligence in Higher Education: Investigating Students' Readiness to Improve Learning Outcomes and Academic Skills
Abstract
The rapid advancement of generative artificial intelligence (GenAI) is transforming teaching and learning practices in higher education. This research investigates students' readiness to use GenAI to improve learning outcomes and develop academic skills using a modified UTAUT2 framework. Survey data collected from 131 higher education students in Serbia were analyzed using PLS-SEM. The results showed that performance expectancy, perceived value, habit, and trust had statistically significant positive effects on behavioral intention to use GenAI, with habit emerging as the strongest predictor. Effort expectancy, perceived risk, social influence, and facilitating conditions did not show statistically significant direct effects. The model explained 90.0% of the variance in behavioral intention, although residual discriminant validity and collinearity concerns should be considered when interpreting individual effects. The findings highlight the importance of established usage habits, perceived educational value, expected performance benefits, and trust in shaping students' readiness to integrate GenAI into the learning process.