The application of judicial review doctrines to automated administration in the United Kingdom

Sinclair, Alexandra Joan (2025) The application of judicial review doctrines to automated administration in the United Kingdom. PhD thesis, London School of Economics and Political Science.

Text - Submitted Version Restricted to Repository staff only until 18 June 2027. Download (2MB)

Abstract

This thesis evaluates the application of English doctrines of judicial review to uses of automation and machine learning in administrative decision-making. The thesis undertakes a close analysis of five doctrines of judicial review that are likely to be engaged by the use of automation and machine learning technologies in government. These doctrines are: reviewability of decisions, the Carltona doctrine, fettering of a discretion, irrelevant considerations and the duty to give reasons. The significant contribution of this thesis is that to date there has not been a granular and sustained treatment of how the grounds of review might apply to automated decision-making in the UK state. This thesis seeks to fill this gap in the literature and contribute to knowledge through an extended examination of the five selected grounds of review. The thesis answers the question: how will these doctrines in their current forms likely apply to uses of automation and machine learning in government? The thesis seeks to examine to what extent these doctrines are capable of meeting their original objectives in an environment of changing administrative behaviour due to uses of automation and machine learning. The thesis argues that some of the grounds of review examined provide a doctrinal foundation which can be flexibly applied to a new automated context and can exercise significant restrictions and constraints over unfair or unlawful uses of automated decision-making in government. Other doctrines demonstrate a more significant clash between what the law currently requires from a decision-maker and whether the technology can satisfy those requirements. By examining these grounds of review in detail I seek to shed a light on the aspects of judicial review’s doctrinal architecture that are and are not readily applicable to uses of automation and machine learning. Through this analysis I lay the foundation for future conversations and research about which areas of review require evolution or where further legal regulation of automation and machine learning is required to respond to this new administrative context.

Item Type:	Thesis (PhD)
Additional Information:	© 2025 Alexandra Joan Sinclair
Library of Congress subject classification:	J Political Science > JF Political institutions (General) K Law > K Law (General) K Law > KD England and Wales Q Science > Q Science (General)
Sets:	Departments > Law
Supervisor:	Poole, Thomas and Lynskey, Orla
URI:	http://etheses.lse.ac.uk/id/eprint/4880

Actions (login required)

Record administration - authorised staff only