Extending the Technology Acceptance Model for Mathematics Education: AI Literacy, AI-TPACK, and Self-Efficacy Among Pre-Service Teachers

This research aims to investigate pre-service mathematics teachers’ (PsMT) intention to use artificial intelligence (AI) technologies. Specifically, this study modified the technology acceptance model (TAM) for this investigation by including variables such as AI literacy, AI-TPACK, and AI self-efficacy. Utilizing a correlation cross-sectional design was the objective of the research. In Türkiye, data were collected from 696 PsMT at 13 universities and analyzed with covariance-based structural equation modeling. The results showed that intention to use AI is primarily determined by perceived usefulness, perceived ease of use, and attitude. Furthermore, AI literacy was less likely to be a predictor of behavioral intention than an antecedent that supports proximal constructs such as perceived usefulness, AI-TPACK, and self-efficacy. AI-TPACK provided support for perceived ease of use and self-efficacy, but not for perceived usefulness or behavioral intention directly. As a consequence, self-efficacy has helped students see AI tools as less intimidating and has cultivated a more positive attitude towards them. The findings indicate that acceptance of AI is a multidimensional process encompassing cognitive, pedagogical, and psychological dimensions, underscoring the importance of comprehensive learning experiences in Teacher Education Institutions that address these dimensions.