Active Fairness Instead of Unawareness

TL;DR

This paper challenges the traditional fairness approach of ignoring sensitive attributes in AI systems, proposing instead to actively use them to monitor and mitigate discrimination in large datasets.

Contribution

It introduces a novel perspective advocating for the active use of sensitive attributes to improve fairness, moving beyond the outdated unawareness paradigm.

Findings

Active use of sensitive attributes can better detect discrimination.

Traditional unawareness methods are ineffective with correlated large datasets.

Proposes a paradigm shift towards fairness through active monitoring.

Abstract

The possible risk that AI systems could promote discrimination by reproducing and enforcing unwanted bias in data has been broadly discussed in research and society. Many current legal standards demand to remove sensitive attributes from data in order to achieve "fairness through unawareness". We argue that this approach is obsolete in the era of big data where large datasets with highly correlated attributes are common. In the contrary, we propose the active use of sensitive attributes with the purpose of observing and controlling any kind of discrimination, and thus leading to fair results.