Lakshmi Balasubramanian, Utkarsh Contractor, & Chris Lemons | Harnessing AI to Enhance Reading Comprehension & Learning Outcomes for Students with Disabilities

Students engaged with science and social studies texts to extract main ideas and answer comprehension questions using AI assistance. The study compared four conditions: baseline, no support, adult support, and AI support, with outcomes measured via rubric-based scoring. Key findings include improved response quality in AI-supported conditions and insights into the design and implementation of effective AI tools in education.

In addition, the session explores the current and future landscape of AI in education, emphasizing the convergence of language intelligence (e.g., large language models like GPT4-series, o3-series, Claude-series, etc..) and multi-modality. It highlights how emerging technologies—combined with audio and visual interfaces —are transforming assistive tools and enabling adaptive, personalized learning with universal design. The discussion will focus on the collaborative potential between educators, engineers, and AI systems to enhance student experiences and improve learning outcomes.