Addressing the void of AI policies in education for students with specific learning disabilities

Abstract

The purpose of this study was to identify the current state of artificial intelligence (AI) policies in U.S. education and propose actionable recommendations through large language model–based topic modeling and Delphi surveys. Out of 12 policy documents released between 2015 and 2025, only two documents (National Center for Learning Disabilities, 2024; W.A. v. Clarksville/Montgomery County School System, 2024) specifically addressed learning disabilities. Policy documents addressing topics such as AI-driven risk assessment, data protection, legal risk management, and ethical guidelines covering other disabilities and general AI in education policy were provided as baselines that could be discussed and validated through the following Delphi surveys involving 17 experts from diverse stakeholder groups. A total of 36 policy items across five thematic categories of inclusive and personalized learning (11 items), ethics, equity, and inclusion (nine items), student empowerment and AI literacy (six items), assessment and research (six items), and educator preparation (four items) were proposed. Based on experts’ ranking of the top 10 policy items important for students, the most essential policy suggestions include student empowerment and AI literacy.