Generative AI Tools and Learning Outcomes for Students with Learning Disabilities in K–12 Education: A Narrative Integrative Review

Abstract

Background: Generative artificial intelligence (GenAI) — including large language models (LLMs), AI chatbots, adaptive tutor systems, and automated feedback tools — is rapidly reshaping discourse in educational research. For students with learning disabilities (LD) — such as dyslexia, dysgraphia, and dyscalculia — these technologies hold theoretical promise for personalized scaffolding, accessible feedback, and self-regulated learning support.
Purpose: This narrative integrative review synthesizes descriptive evidence on the potential and limitations of GenAI and related AI tools on learning outcomes for K–12 students with LD, situating findings within cognitive, metacognitive, and inclusive education frameworks.
Methods: Evidence was purposively selected for descriptive synthesis based on relevance to the topic, recency (primarily 2017–2025), methodological quality, and focus on adaptive or generative AI tools in educational contexts. Peer-reviewed research, high-quality conceptual pieces, and reputable reports were included.
Results: Direct empirical research specifically on GenAI tools with K–12 LD populations is currently limited. Evidence for related adaptive AI systems — including intelligent tutoring systems (ITS) and automated feedback platforms — indicates potential positive effects on reading, writing, and mathematics performance, improved self-regulated learning behaviors, and increased engagement. Mechanistically, AI supports cognitive load reduction, personalized feedback loops, and multimodal accessibility, although risks such as dependency and accuracy concerns persist.
Conclusions: While GenAI offers promising affordances for students with LD, rigorous empirical research in K–12 special education settings is urgently needed. Future work should prioritize controlled intervention trials, standardized outcome measures, and inclusive co-design with learning disability communities.