Artificial Intelligence in Inclusive and Special Education: A Narrative Survey

Abstract

Background: Artificial intelligence (AI) is increasingly discussed as a means of supporting inclusive and special education through adaptive learning, assistive technologies, learning analytics, automated feedback, and generative AI. However, the educational value of AI for learners with disabilities depends on accessibility, pedagogical alignment, data protection, teacher agency, and equitable implementation. Objectives: This narrative survey synthesizes current literature and policy guidance on AI in inclusive and special education, with emphasis on applications, opportunities, risks, and responsible implementation. Methods: The paper uses a structured narrative review approach based on peer-reviewed review articles, foundational AI-in-education literature, and international policy sources relevant to inclusive education, assistive technology, and AI governance. Results: The synthesis indicates that AI may support personalization, accessibility, communication, teacher workload reduction, early identification of support needs, and formative feedback. At the same time, evidence remains uneven across disability groups, educational levels, and low-resource contexts. Major risks include inaccurate outputs, bias, privacy concerns, limited transparency, over-reliance, insufficient teacher preparation, and inequitable access. Conclusion: AI should be positioned as a human-supervised support system rather than a substitute for teachers, special educators, therapists, or institutional responsibility. Responsible adoption requires Universal Design for Learning, learner and family participation, rigorous evaluation, transparent policies, and safeguards for privacy, fairness, and academic integrity.