Innovating Reading Materials Development Based on Generative AI Within the Backward Design Framework

Abstract

Tailoring reading textbooks for personalized learning is essential. However, this process is time-consuming, labor-intensive, and costly for language teachers. This study fills this gap using generative AI to develop customized reading materials based on the backward design framework, which can assess five reading subskills in the teaching objectives. 288 EFL young learners were invited to respond to the reading materials generated by generative AI within 45 minutes. The teaching materials developed by AI were evaluated within the backward design framework. The result found that generative AI could develop customized reading materials for the desired reading subskills. Psychometric models revealed that the difficulties of reading materials were adaptive to students’ reading ability holistically and that the reading materials were unbiased toward different genders. Besides, English teachers believed that AI-generated reading materials were useful and helpful to language teaching and learning. This study provides a new method for developing personalized reading materials for language teaching, learning, and testing using generative AI, which has the potential to increase the usefulness and diversity of teaching materials efficiently.