This study aims to explore the key determinants influencing university students’ behavioural intention to use Generative Artificial Intelligence (BI_GAI) tools in educational settings, as well as the impact of this intention on student satisfaction (SS). Grounded in the Technology Acceptance Model (TAM), the research incorporates the traditional constructs of Perceived Ease of Use (PE) and Perceived Usefulness (PU), and extends the model by integrating Perceived Intelligence (PI), Perceived Trust (PT), Perceived Risk (PR), Expected Benefits (EB), and Technology Self-Efficacy (TSE).

Data were collected from 775 students at a Portuguese higher education institution through a questionnaire comprising 40 items across nine constructs (PE, PI, PU, PT, PR, BI_GAI, EB, TSE, and SS), alongside sociodemographic variables.

The data were analysed using Partial Least Squares Structural Equation Modelling (PLS-SEM). The results reveal that PE and PI have a significant positive effect on Behavioural Intention to Use GAI (BI_GAI), whereas PU does not have a statistically significant direct influence. Perceived Trust (PT) emerges as a key mediating variable in the relationship between PU and BI_GAI, while Perceived risk (PR) does not act as a significant mediator between the TAM constructs and BI_GAI. Behavioural Intention to Use GAI has the strongest direct influence on Student Satisfaction (SS), highlighting its central role in understanding students’ engagement with GAI tools. Moreover, both EB and TSE significantly affect SS, both directly and indirectly through BI_GAI.

These findings support the development of an expanded TAM-based model that provides a more holistic perspective on the technological, psychological, and educational factors shaping GAI adoption in higher education. The inclusion of constructs such as perceived intelligence, trust, and technology self-efficacy offers deeper insights into the mechanisms through which students evaluate and adopt GAI for learning purposes, ultimately contributing to enhanced academic satisfaction.

Keywords: University students’ perceptions, generative artificial intelligence, data analysis, PLS-SEM.