Lessons Learned: Surveying the Practicality of Differential Privacy in the Industry

Abstract

Since its introduction in 2006, differential privacy has emerged as a predominant statistical tool for quantifying data privacy in academic works. Yet despite the plethora of research and open-source utilities that have accompanied its rise, with limited exceptions, differential privacy has failed to achieve widespread adoption in the enterprise domain. Our study aims to shed light on the fundamental causes underlying this academic-industrial utilization gap through detailed interviews of 24 privacy practitioners across 9 major companies. We analyze the results of our survey to provide key findings and suggestions for companies striving to improve privacy protection in their data workflows and highlight the necessary and missing requirements of existing differential privacy tools, with the goal of guiding researchers working towards the broader adoption of differential privacy. Our findings indicate that analysts suffer from lengthy bureaucratic processes for requesting access to sensitive data, yet once granted, only scarcely-enforced privacy policies stand between rogue practitioners and misuse of private information. We thus argue that differential privacy can significantly improve the processes of requesting and conducting data exploration across silos, and conclude that with a few of the improvements suggested herein, the practical use of differential privacy across the enterprise is within striking distance.

Don't Look at the Data! How Differential Privacy Reconfigures the Practices of Data Science

“…Concurrent work by Munilla Garrido et al [27] investigates the 'academic-industrial DP utilization gap' through interviews with data analysts and data stewards in major companies that have not yet deployed DP. Their findings about the barriers to adoption of DP, as well as the promises of using DP, support the findings in this paper.…”

Geometrically-aggregated training samples: Leveraging summary statistics to enable healthcare data democratization

“…By integrating differential privacy techniques into the data generation process, synthetic datasets can be created with privacy assurances (10; 11; 13; 21). However, it's important to note that although differential privacy provides statistical privacy guarantees, the addition of significant noise can reduce the usefulness of the data for ML development, especially when the original data itself may be inherently noisy (22). Moreover, the absence of a standardized approach or metric for implementing differential privacy in ML poses challenges in terms of implementation, evaluation, and practical deployment.…”

Exploring the Privacy Risks of Adversarial VR Game Design

“…Because virtual reality devices generate a stream of data relating to the motion of a user, and people are known to subconsciously reveal information about themselves via their motion, it is natural to question the extent to which personal attributes are inferable in VR. According to popular literature reviews of the field [10,14,16,19,23,24,34,39,42,50,61,70], the vast majority of existing VR privacy research has focused on passive observation.…”

“…It has long been understood that individuals exhibit distinct biomechanical motion patterns that can be used to identify them or infer their personal attributes [13,27,31,33,48,53], which researchers have shown can be exploited to identify and profile users in VR [23,35,41,44]. While existing work has largely focused on passive observation of VR users, the success of games specifically designed to harvest user data [25] on conventional social platforms motivates us to investigate similar active attacks in VR.…”