Explainable AI Systems Must Be Contestable: Here's How to Make It Happen

TL;DR

This paper defines contestability in explainable AI, proposes a comprehensive framework and assessment scale, and demonstrates how to implement and evaluate contestability in real-world AI systems to meet regulatory standards.

Contribution

It provides the first formal definition of contestability in XAI, a modular framework for implementation, and a practical assessment scale validated through case studies.

Findings

State-of-the-art systems often lack true contestability.

The framework identifies specific gaps in current AI systems.

The assessment scale enables targeted improvements in contestability.

Abstract

As AI regulations around the world intensify their focus on system safety, contestability has become a mandatory, yet ill-defined, safeguard. In XAI, "contestability" remains an empty promise: no formal definition exists, no algorithm guarantees it, and practitioners lack concrete guidance to satisfy regulatory requirements. Grounded in a systematic literature review, this paper presents the first rigorous formal definition of contestability in explainable AI, directly aligned with stakeholder requirements and regulatory mandates. We introduce a modular framework of by-design and post-hoc mechanisms spanning human-centered interfaces, technical architectures, legal processes, and organizational workflows. To operationalize our framework, we propose the Contestability Assessment Scale, a composite metric built on more than twenty quantitative criteria. Through multiple case studies across diverse application domains, we reveal where state-of-the-art systems fall short and show how our framework drives targeted improvements. By converting contestability from regulatory theory into a practical framework, our work equips practitioners with the tools to embed genuine recourse and accountability into AI systems.