Big data, AI, and proxy discrimination: a review of Professor Margarida Lima Rego’s ‘dissent’ on insurance discrimination bans

Stevens, Madison (2022) Big data, AI, and proxy discrimination: a review of Professor Margarida Lima Rego’s ‘dissent’ on insurance discrimination bans. Journal of the British Insurance Law Association, Platinum Editio.

Record type: Article

Abstract

In July 2021, Professor Margarida Lima Rego delivered a presentation on her dissent from discrimination bans in insurance. Since the 2011 ruling of the Court of Justice of the European Union in Test Achats, it has been settled that the use of statistical discrimination within the European Union is prohibited. Dissenting from this approach, Professor Lima Rego proposes a more nuanced approach to statistical discrimination. Instead, she advances a twofold test to distinguish between admissible and inadmissible forms of discrimination. Within her dissent, Professor Lima Rego notes how the rise of AI and big data is changing insurance practice and shifting questions about the acceptability of statistical discrimination. This paper will provide a review of Professor Lima Rego’s dissent and elaborate on her claims that AI and big data are changing the kinds of discrimination we see in insurance. Specifically, it will explain the concept of “proxy discrimination” which is likely to occur when machine-learning algorithms are used. This kind of proxy discrimination is exceedingly difficult to detect and manage under discrimination law. In recognition of this, this paper will explore whether the balanced approach proposed by Professor Lima Rego would be more appropriate to regulate the changing nature of insurance discrimination.

This record has no associated files available for download.