Configurational effects of personal innovativeness, self-efficacy, and perceived risk on AI adoption in media students

Generative artificial intelligence (G-AI) is driving systemic transformations in the media industry, redefining the cultivation paradigm of media majors in higher education. This shift necessitates that future media professionals integrate AI thinking and master the application of digital intelligence technologies. To address the knowledge gap in understanding media students' adoption of AI technologies, this study extends the classical Technology Acceptance Model by incorporating personal innovativeness, AI self-efficacy, and perceived risk, thereby constructing a new and comprehensive model. A survey of 588 media students was conducted, and the data were analyzed using PLS-SEM and fsQCA. The results indicate that: (1) personal innovativeness and AI self-efficacy significantly enhance both perceived usefulness and perceived ease of use; (2) perceived usefulness and perceived ease of use are core determinants of AI behavioral intention, mediating the effects of personal innovativeness and AI self-efficacy; (3) perceived risks can negatively moderate the effect of perceived usefulness and perceived ease of use on AI behavioral intention; (4) AI behavioral intention is shaped not by a single linear determinant but by intersecting configurations of factors, leading to three distinct enhancement pathways: self-driven, efficacy-oriented, and risk-resistant. This study advances the Technology Acceptance Model by integrating psychological and risk-based constructs, offering theoretical insights into the multidimensional driving forces of AI adoption and practical strategies for fostering AI competencies in media students.