Assessing Model-Agnostic XAI Methods against EU AI Act Explainability Requirements
For practitioners and regulators, this work provides a practical tool to bridge XAI methods and legal compliance, though it is an incremental step without empirical validation.
The paper addresses the gap between existing XAI methods and EU AI Act requirements by proposing a qualitative-to-quantitative scoring framework that maps model-agnostic XAI interpretability features to regulatory compliance scores, helping practitioners identify suitable XAI solutions for legal explanation requirements.
Explainable AI (XAI) has evolved in response to expectations and regulations, such as the EU AI Act, which introduces regulatory requirements on AI-powered systems. However, a persistent gap remains between existing XAI methods and society's legal requirements, leaving practitioners without clear guidance on how to approach compliance in the EU market. To bridge this gap, we study model-agnostic XAI methods and relate their interpretability features to the requirements of the AI Act. We then propose a qualitative-to-quantitative scoring framework: qualitative expert assessments of XAI properties are aggregated into a regulation-specific compliance score. This helps practitioners identify when XAI solutions may support legal explanation requirements while highlighting technical issues that require further research and regulatory clarification.