AI anxiety and adoption intention in higher education based on an extended TAM-UTAUT and PLS-SEM analysis

This study aims to deconstruct the complex relationship between artificial intelligence anxiety and generative AI adoption intention in the context of higher education. An extended analytical framework is constructed by integrating the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT). Taking faculty and students from three universities in Sichuan Province as the research objects, data are collected through a questionnaire survey, and empirical research is conducted using the Partial Least Squares Structural Equation Modeling (PLS-SEM). Focusing on three dimensions of AI anxiety—AI learning anxiety, AI sociotechnical blindness anxiety, and AI job displacement anxiety—the study systematically examines their indirect impact paths on adoption intention through mediating variables such as performance expectancy and effort expectancy, and investigates the moderating effects of disciplinary background and AI self-efficacy. The findings reveal that AI anxiety exhibits a significant “double-edged sword” effect on generative AI adoption intention: AI job displacement anxiety comprehensively inhibits the core variables of technology acceptance; in contrast, AI sociotechnical blindness anxiety not only positively promotes effort expectancy and adoption intention but also negatively affects performance expectancy. In the technology acceptance mechanism, effort expectancy and social influence are the factors driving adoption intention, while disciplinary background and AI self-efficacy regulate the transmission paths of anxiety by shaping cognitive paradigms and psychological resources. The theoretical value of this study lies in breaking through the simplistic inhibition hypothesis of emotional factors in technology acceptance research. Practically, it provides empirical evidence for higher education institutions to formulate differentiated AI training strategies.