Who Benefits from AI Explanations? Towards Accessible and Interpretable Systems
It addresses the problem of making AI explanations accessible for users with vision impairments, which is an incremental step in XAI usability.
This paper investigates accessibility gaps in eXplainable AI (XAI) for users with vision impairments, finding through a literature review that evaluations rarely include disabled users and most explanations rely on visual formats, and presenting a methodological proof of concept that suggests simplified explanations and multimodal presentation improve comprehensibility.
As AI systems are increasingly deployed to support decision-making in critical domains, explainability has become a means to enhance the understandability of these outputs and enable users to make more informed and conscious choices. However, despite growing interest in the usability of eXplainable AI (XAI), the accessibility of these methods, particularly for users with vision impairments, remains underexplored. This paper investigates accessibility gaps in XAI through a two-pronged approach. First, a literature review of 79 studies reveals that evaluations of XAI techniques rarely include disabled users, with most explanations relying on inherently visual formats. Second, we present a four-part methodological proof of concept that operationalizes inclusive XAI design: (1) categorization of AI systems, (2) persona definition and contextualization, (3) prototype design and implementation, and (4) expert and user assessment of XAI techniques for accessibility. Preliminary findings suggest that simplified explanations are more comprehensible for non-visual users than detailed ones, and that multimodal presentation is required for more equitable interpretability.