Mathematics presents a unique challenge in the field of Artificial Intelligence in Education, as many algebraic notations and graphical representations are not easily interpretable by language models. Despite the growing integration of AI in education, the mathematics education community has yet to adequately address the foundational tasks of defining and developing pedagogically sound datasets with rigorous benchmarks. Many AI-powered resources, including AI tutors, are developed by large language model vendors, whose priorities are often driven more by market trends and hype than by learning sciences (e.g., addressing the needs and characteristics of learners and teachers). Consequently, some inappropriately designed AI tutors may reinforce learners’ misconceptions and hinder their critical thinking and problem-solving skills. In this paper, we present findings from a small-scale interdisciplinary research study that explored maths-centric Intelligent Tutoring Systems (ITS), with a particular focus on reviewing available training databases. We also present an interdisciplinary model for developing effective maths-centric ITS from a sociocultural lens.

Keywords: Intelligent tutoring system (ITS), mathematics education, AIED, machine learning, sociocultural theory.