A multidimensional analysis of generative AI adoption in academia: synthesizing TAM3 and UTAUT2 frameworks to predict students’ attitudes and learning effectiveness

Abstract

The rapid integration of generative artificial intelligence (GenAI) tools into higher education raises important questions regarding their impact on student learning. Based on established technology acceptance frameworks, this study develops and empirically tests an integrative model that explains students’ attitudes toward GenAI tools and their perceived learning effectiveness. Data were collected from 566 undergraduate students using a structured questionnaire. The main findings indicate that prior experience, perceived merits, and training in using GenAI tools are positively associated with and constitute strong predictors of both attitudes of students toward GenAI use and perceived learning effectiveness, while ethical knowledge and technological orientation do not emerge as significant predictors. Perceived drawbacks are also significant, suggesting that ambivalence and concerns about GenAI can diminish students’ positive attitudes and reduce their perceived usefulness. Differences across gender further suggest that technology adoption processes are not uniform but contingent upon user characteristics. Theoretically, this study contributes to the literature by extending technology acceptance models into the domain of GenAI and by linking attitudinal constructs with learning-related outcomes within a unified framework. Practically, the findings suggest that higher education institutions should not only provide access to GenAI tools but also complement this access with pedagogical support, including structured training, comprehensive ethical guidance, and the development of students’ GenAI literacy. Overall, the study provides a nuanced understanding of how GenAI tools are adopted and leveraged in academic contexts and offers a foundation for future research on GenAI-enabled learning.