Lui, A, Durodola, L and Lamb, G A Right to Explanation for Algorithmic Credit Decisions in the UK. Law, Innovation and Technology. ISSN 1757-9961 (Accepted)

Text A Right to Explanation for Algorithmic Credit Decisions in the UK.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (407kB)

Abstract

This article argues for a statutory right to explanation in automated credit decision-making in the UK, as transparency and accountability are central to the rule of law. This is based on two premises. First, from a moral standpoint, we demonstrate that there is a double level of distrust in financial services and algorithms. Algorithms are unpredictable and can make unreliable, strange decisions. Algorithmic challenges such as bias, discrimination and unfairness are exacerbated by the opacity problem commonly known as the ‘black box’ phenomenon. The informed consent process in automated credit decision-making is thus incomplete, which requires an ex-post right to explanation for completing the informed consent procedure. Secondly, our doctrinal and comparative legal methodologies reveal that countries such as the USA, Canada, European Union, China and Poland already provide a right to explanation to credit applicants under certain circumstances. Central to our argument is the introduction of new empirical evidence of public surveys regarding a desire from the public to have a right to explanation for unsuccessful credit applications. We argue for a statutory right to meaningful and accessible local feature-based information to automated credit decision making, which should include objective criteria and weightings used by banks.

Item Type:	Article
Uncontrolled Keywords:	1801 Law
Subjects:	K Law > K Law (General) T Technology > T Technology (General)
Divisions:	Law
Publisher:	Taylor and Francis Group
SWORD Depositor:	A Symplectic
Date Deposited:	06 Jan 2025 12:04
Last Modified:	06 Jan 2025 12:15
URI:	https://researchonline.ljmu.ac.uk/id/eprint/25169

View Item