Artificial intelligence in web accessibility: A systematic mapping study
Milton Campoverde-Molina, Sergio Luján-Mora
Computer Standards & Interfaces, 96, 104055, p. 1-29, 2026. ISSN: 0920-5489. https://doi.org/10.1016/j.csi.2025.104055
(CSI'26)
Revista / Journal
JCR IF (2024): 3.1 - Computer Science, Software Engineering: 41/128 (Q2); Computer Science, Hardware & Architecture: 21/60 (Q2)
Resumen
The popularization and new renaissance of artificial intelligence (AI) have inspired the creation of new applications that help developers improve website accessibility that benefits users with and without disabilities. Therefore, this research presents a systematic mapping study (SMS) because AI in web accessibility has been gaining more interest nowadays with the exposure of works that require an SMS to systematize and consolidate the literature. Through a literature review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), 53 studies from ACM Digital Library, IEEE Xplore, Scopus, and Web of Science were identified for inclusion in this review. The main results of this SMS are APIs with AI, web applications and plugins with AI, image and voice recognition with AI, limitations and challenges of AI in web accessibility, correction and testing of web accessibility with AI, automatic correction of web accessibility with AI, web navigation with conversational agents with AI, web and mobile application accessibility with AI, good practices in web accessibility for AI, accessibility of web forms and elements with AI. According to the results, in the studies, alternative texts were created for the images of the websites, AI helped generate accessible HTML code using well-defined prompts, AI tools must comply with Web Content Accessibility Guidelines (WCAG), machine learning was the most used approach, the most used language models were large language models (LLM) and accessibility barrier correction using ChatGPT. The primary contribution of this SMS lies in its analysis of the current state of AI research related to web accessibility and the identification of trends and gaps in this research area. This SMS is intended for researchers, programmers, and software development companies that may use language models, AI tools, or emerging technologies in web accessibility to mitigate website accessibility barriers.