Assessing AI-TPACK readiness in mathematics teacher education: The role of self-efficacy and teaching beliefs

As artificial intelligence (AI) becomes integral to contemporary classrooms, mathematics teacher education students (MTES) must develop AI-technological pedagogical content knowledge (AI-TPACK) to effectively orchestrate intelligent tools. This study first developed and validated a domain-specific AI-TPACK instrument tailored to the unique pedagogical and logical demands of mathematics education. A survey of 412 Chinese MTES revealed that their AI-TPACK readiness is currently at a preliminary stage, with no significant differences observed across grade levels within the teacher education curriculum. Furthermore, a structural equation model (SEM) was employed to investigate the correlational pathways between knowledge domains and psychological factors. The results demonstrate that while self-efficacy serves as a significant positive predictor of AI-TPACK, strong traditional teaching beliefs can paradoxically act as a cognitive barrier to AI integration. These findings provide empirical evidence for the necessity of redesigning mathematics teacher training programs to address both technical proficiency and psychological readiness (defined as the synergy between self-efficacy and pedagogical beliefs).