The Limits of Differential Privacy (and its Misuse in Data Release and Machine Learning)

TL;DR

This paper critically examines the limitations of differential privacy, emphasizing that it is not a universal solution and highlighting risks of its misuse in data collection, release, and machine learning contexts.

Contribution

It provides a comprehensive review of differential privacy's limitations and warns against its inappropriate application beyond its intended scope.

Findings

DP is not a universal privacy solution

Misuse of DP can lead to privacy breaches

Limitations of DP in data release and ML contexts

Abstract

Differential privacy (DP) is a neat privacy definition that can co-exist with certain well-defined data uses in the context of interactive queries. However, DP is neither a silver bullet for all privacy problems nor a replacement for all previous privacy models. In fact, extreme care should be exercised when trying to extend its use beyond the setting it was designed for. This paper reviews the limitations of DP and its misuse for individual data collection, individual data release, and machine learning.