Integration of MATH41 and Generative AI in Pre-Service Mathematics Teacher Education: An Empirical Study on Lesson Design Competency
Sejun Oh
Abstract
Generative artificial intelligence (AI) tools are rapidly transforming mathematics education by enabling automated problem generation, dynamic visualizations, and adaptive learning experiences. This study presents an empirical study on incorporating MATH41 into a 15-week pre-service teacher preparation course for mathematics majors. Twenty-four participants learned to create parameterized math problems, generate vector graphics. Following guided training, each participant designed and micro-taught a 20-minute lesson incorporating MATH41-generated resources. Reflection journals and final lesson plans were analyzed thematically, while 20 participants completed a self-assessment survey on lesson design competency. Results reveal that the automated problem generation capabilities motivated pre-service teachers to explore a broader range of instructional strategies, including personalized tasks and diverse problem variants. Reflection data indicate that while integrating AI tools can significantly boost confidence and creativity in lesson planning, careful pedagogical alignment remains essential to avoid superficial learning. Participants underscored the importance of maintaining teacher oversight—especially when adapting AI-generated problems for particular learner needs. Additionally, their post-course self-assessments showed high confidence in digital tool integration, yet they acknowledged that anticipating student misconceptions requires further field experience. Overall, this study contributes to understanding how teacher education programs can enhance lesson design competencies via structured AI tool integration. It also highlights the critical role of reflective practice in ensuring that automated content creation fosters deeper instructional effectiveness rather than uncritical AI dependency.