Student-designed educational resources for Lean: exploring ChatGPT and Retrieval-Augmented Generation

Abstract Recent technological developments are rapidly changing the teaching and learning of mathematics. Among such tools, we focus on Lean and ChatGPT, an area which remains under-researched in mathematics education. The aim of this study is to explore university students’ experience and views of creating educational resources using Lean and ChatGPT. Drawing on the framework of documentation work, we analyse the resources created by four undergraduate and one postgraduate student, who had previously used Lean and were invited to design materials to support future students learning Lean. Through an analysis of their customised ChatGPT tools using Retrieval-Augmented Generation, educational resources, and interview data, we show how students retrospectively described developing a more critical stance towards Lean and ChatGPT through their design work. Our findings highlight the diversity of documentation work with ChatGPT, shedding light on the complexities of integrating AI into mathematics education.