Header

UZH-Logo

Maintenance Infos

The chilling effects of algorithmic profiling: mapping the issues


Büchi, Moritz; Fosch-Villaronga, Eduard; Lutz, Christoph; Tamò-Larrieux, Aurelia; Velidi, Shruthi; Viljoen, Salome (2020). The chilling effects of algorithmic profiling: mapping the issues. Computer Law & Security Review (CLSR), 36:105367.

Abstract

In this article, we provide an overview of the literature on chilling effects and corporate profiling, while also connecting the two topics. We start by explaining how profiling, in an increasingly data-rich environment, creates substantial power asymmetries between users and platforms (and corporations more broadly). Inferences and the increasingly automated nature of decision-making, both based on user data, are essential aspects of profiling. We then connect chilling effects theory and the relevant empirical findings to corporate profiling. In this article, we first stress the relationship and similarities between profiling and surveillance. Second, we describe chilling effects as a result of state and peer surveillance, specifically. We then show the interrelatedness of corporate and state profiling, and finally spotlight the customization of behavior and behavioral manipulation as particularly significant issues in this discourse. This is complemented with an exploration of the legal foundations of profiling through an analysis of European and US data protection law. We find that while Europe has a clear regulatory framework in place for profiling, the US primarily relies on a patchwork of sector-specific or state laws. Further, there is an attempt to regulate differential impacts of profiling via anti-discrimination statutes, yet few policies focus on combating generalized harms of profiling, such as chilling effects. Finally, we devise four concise propositions to guide future research on the connection between corporate profiling and chilling effects.

Abstract

In this article, we provide an overview of the literature on chilling effects and corporate profiling, while also connecting the two topics. We start by explaining how profiling, in an increasingly data-rich environment, creates substantial power asymmetries between users and platforms (and corporations more broadly). Inferences and the increasingly automated nature of decision-making, both based on user data, are essential aspects of profiling. We then connect chilling effects theory and the relevant empirical findings to corporate profiling. In this article, we first stress the relationship and similarities between profiling and surveillance. Second, we describe chilling effects as a result of state and peer surveillance, specifically. We then show the interrelatedness of corporate and state profiling, and finally spotlight the customization of behavior and behavioral manipulation as particularly significant issues in this discourse. This is complemented with an exploration of the legal foundations of profiling through an analysis of European and US data protection law. We find that while Europe has a clear regulatory framework in place for profiling, the US primarily relies on a patchwork of sector-specific or state laws. Further, there is an attempt to regulate differential impacts of profiling via anti-discrimination statutes, yet few policies focus on combating generalized harms of profiling, such as chilling effects. Finally, we devise four concise propositions to guide future research on the connection between corporate profiling and chilling effects.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Downloads

3 downloads since deposited on 05 Feb 2020
3 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:06 Faculty of Arts > Department of Communication and Media Research
Dewey Decimal Classification:700 Arts
Uncontrolled Keywords:Profiling, inferences, chilling effects, surveillance, data protection, privacy, algorithms, big data, digital footprints
Language:English
Date:1 April 2020
Deposited On:05 Feb 2020 10:46
Last Modified:29 Mar 2020 01:05
Publisher:Elsevier
ISSN:0267-3649
OA Status:Closed
Publisher DOI:https://doi.org/10.1016/j.clsr.2019.105367

Download

Closed Access: Download allowed only for UZH members

Content: Accepted Version
Language: English
Filetype: PDF - Registered users only until 22 November 2021
Size: 326kB
View at publisher
Embargo till: 2021-11-22