Ummu Ajirah Abdul Rauf, Mazzlida Mat Deli, Nanang Husin
The purpose of this study is to develop an AI-based, inclusive e-learning model to reduce the educational risks faced by disabled students. By exploring the potential of AI to support equitable participation in digital learning, this research addresses remaining challenges, including insufficiently adaptable content, social isolation, and varying degrees of access. Using a qualitative multi-case approach, data are drawn from a literature review, interviews, and feedback obtained during focus group discussions with students who have visual, hearing, physical, and cognitive disabilities at three Malaysian universities. Thematic analysis in NVivo found that AI-supported technologies, such as automated captioning, text-to-speech, and predictive learning analytics, support listening comprehension, collaboration, and engagement; however, challenges remain regarding algorithmic bias, privacy, and infrastructure readiness. This led to the development of the AI-Driven Inclusive E-Learning Risk-Mitigation Framework (AI-IERMF), which embraces Universal Design for Learning (UDL) principles for real-time adaptive analytics. Theoretically, this study advances the social model of disability by adopting a predictive risk approach. Practically, it helps policymakers, educators, and developers create transparent, ethically sensitive, and resource-aware e-learning systems. Overall, the study informs inclusive education and shifts accessibility from reactive compliance to proactive digital empowerment. © 2025, Intellectual Research and Development Education Foundation (YRPI). All rights reserved.
Graduate School of Business, Universiti Kebangsaan Malaysia, Malaysia; Faculty of Economics Department, State University of Surabaya, Indonesia
