How teachers navigate adoption of Generative AI in technical and vocational education and training: China's evidence from configurational analysis.

Applying generative artificial intelligence (GAI) into technical and vocational education and training (TVET) faces both opportunities and risks, demanding an in-depth understanding of how teachers navigate the technology adoption. This study renews the Unified Theory of Acceptance and Use of Technology (UTAUT) model through the inclusion of three variables (perceived risk, perceived trust and digital literacy), examining the configurational pathways influencing TVET teachers' intentions to utilize GAI. Analyzing the survey data from 400 respondents in China, an approach to combining fuzzy-set Qualitative Comparative Analysis (fsQCA) and Necessary Condition Analysis (NCA) was used. This study finds that performance expectancy serves as a necessary condition for high-level willingness of GAI adoption; three distinct and equifinal causal pathways are identified (risk-indifferent, risk-tolerant and risk-cogovernance) and cooperatively lead to high behavioral intention. The contributions at the theoretical, methodological, and practical levels made by this study can advance existing research on similar agendas and offer insights for constructing a collaborative governance mechanism for GAI risks in TVET.