A Survey on Methods and Metrics for the Assessment of Explainability under the Proposed AI Act

TL;DR

This paper reviews explainability metrics in AI, emphasizing their role in compliance with the proposed EU AI Act, and proposes key requirements for effective, standardized explainability measurement.

Contribution

It provides an interdisciplinary analysis of explainability metrics aligned with the AI Act, proposing essential features for future metrics to ensure compliance and standardization.

Findings

Metrics should be risk-focused and model-agnostic

Explainability metrics need to be goal-aware and accessible

Current metrics partially meet proposed requirements

Abstract

This study discusses the interplay between metrics used to measure the explainability of the AI systems and the proposed EU Artificial Intelligence Act. A standardisation process is ongoing: several entities (e.g. ISO) and scholars are discussing how to design systems that are compliant with the forthcoming Act and explainability metrics play a significant role. This study identifies the requirements that such a metric should possess to ease compliance with the AI Act. It does so according to an interdisciplinary approach, i.e. by departing from the philosophical concept of explainability and discussing some metrics proposed by scholars and standardisation entities through the lenses of the explainability obligations set by the proposed AI Act. Our analysis proposes that metrics to measure the kind of explainability endorsed by the proposed AI Act shall be risk-focused, model-agnostic, goal-aware, intelligible & accessible. This is why we discuss the extent to which these requirements are met by the metrics currently under discussion.