Comparing Gen AI adoption in pre- and in-service mathematics teachers

Source abstract

This study examines Generative Artificial Intelligence (Gen AI) integration in a Swedish mathematics education context by investigating readiness, usage patterns, and approaches among pre-service (N=17) and in-service (N=22) mathematics teachers. Using a mixed-methods approach within a teacher training program in spring 2025, the research addresses three questions: (1) current Gen AI readiness levels, behavioral intentions, and attitudes; (2) differences between groups in integration approaches; and (3) practical applications, challenges, and coping strategies. Data collection included quantitative readiness measures and qualitative responses about experiences. Results revealed that both career stage and usage frequency influence Gen AI adoption: career stage shapes the qualitative nature of integration (purposes, approaches, sophistication), while usage frequency emerges as a stronger predictor of learning changes and readiness levels. Pre-service teachers demonstrated higher usage for learning, while in-service teachers showed selective, professionally-informed usage despite similar readiness levels. High-frequency users reported significantly greater learning changes and higher readiness scores, highlighting hands-on experience’s role in developing AI competencies. The study identifies two tentative patterns: pre-service teachers’ developmental integration as learning scaffolds versus in-service teachers’ selective, professionally-informed integration with more sophisticated strategies. These exploratory findings offer preliminary evidence of how mathematics educators at different career stages conceptualize Gen AI integration, and point to directions for future research, targeted professional development and teacher training programs.