The psychology of AI adoption in education: University students' intentions to use Large Language Models for learning from a TAM and TPB perspective

Abstract

The rapid rise of artificial intelligence (AI) and the exponential growth of AI-based tools, particularly LLMs, have spurred extensive global research in recent years. In the education sector, LLMs have demonstrated their potential to enhance students' learning experiences and contribute to improved academic performance. In Vietnam, while recent studies have explored university students' perceptions of using LLMs in learning, there is a notable research gap regarding their actual adoption of these tools for educational purposes. Therefore, this research was conducted to address this gap by investigating the intentions to adopt LLMs for learning by university students, applying the extended Technology Adoption Model (TAM) and Theory of Planned Behavior (TPB) models. The study employed a non-probability convenience sampling approach to gather data from 226 university students in Vietnam who have prior experiences in adapting LLMs for their learning. Partial Least Square Structural Equation Modeling (PLS-SEM) was employed to test the hypotheses and to evaluate the research model. The findings revealed that perceived ease of use, perceived usefulness, and perceived trust in LLMs significantly and positively influence students' attitudes toward adoption. Moreover, all TPB constructs—attitude, subjective norms, and perceived behavioral control—positively impact intention to use LLMs, with subjective norms having the strongest effect. These findings underscore the critical role of social influence in shaping students' adoption behavior within this cultural context. This study contributes to theory by demonstrating the importance of incorporating cultural elements, such as subjective norms, into adoption models. Practically, it provides clear guidance for educators and policymakers to promote LLM integration through peer-led strategies and the cultivation of positive social norms around AI use in education. Ultimately, the study highlights that fostering a supportive and socially endorsed environment is key to encouraging effective adoption of AI technologies in learning. • Explores LLM adoption among Vietnamese university students, addressing a gap in AI education research. • Applies extended TAM and TPB to analyze factors shaping LLM adoption in higher education. • Finds ease of use, usefulness, and trust drive students’ attitudes toward adopting LLMs. • Reveals subjective norms most strongly influence students’ intentions to use LLMs in learning. • Provides insights for educators and policymakers to enhance engagement with AI learning tools.

Topics & Concepts