Can Requirements Engineering Support Explainable Artificial Intelligence? Towards a User-Centric Approach for Explainability Requirements
Umm-e-Habiba, Justus Bogner, Stefan Wagner
TL;DR
The paper addresses the lack of user-centered explainability in AI systems and argues that Requirements Engineering (RE) can bridge the gap between diverse stakeholders and ML developers. It clarifies the distinction between interpretability and explainability, reviews existing RE-XAI synergies, and identifies key challenges such as the absence of a mediator role, definitional ambiguity, stakeholder-centric gaps, and vocabulary fragmentation. A mediator-led RE framework is proposed, with steps to identify stakeholders, elicit and rationalize explainability requirements, establish a common vocabulary, negotiate trade-offs, and classify requirements by stakeholder. The contribution offers a practical direction for integrating explainability as a non-functional requirement in AI development, enabling better alignment among ML engineers, domain experts, and end-users, and setting the stage for more usable and trustworthy AI systems.
Abstract
With the recent proliferation of artificial intelligence systems, there has been a surge in the demand for explainability of these systems. Explanations help to reduce system opacity, support transparency, and increase stakeholder trust. In this position paper, we discuss synergies between requirements engineering (RE) and Explainable AI (XAI). We highlight challenges in the field of XAI, and propose a framework and research directions on how RE practices can help to mitigate these challenges.