Good Administration in AI-Enhanced Banking Supervision: A Risk-Based Approach

67 Pages Posted: 27 Apr 2023 Last revised: 28 Oct 2024

See all articles by Alessio Azzutti

Alessio Azzutti

University of Glasgow - School of Law; National University of Singapore (NUS) - Centre for Banking and Finance Law (CBFL); Institute of Law & Economics - University of Hamburg; European Banking Institute

Pedro Magalhães Batista

University of Leeds; New York University School of Law

Wolf-Georg Ringe

University of Hamburg - Institute of Law and Economics; University of Oxford - Faculty of Law; European Corporate Governance Institute (ECGI)

Date Written: April 27, 2023

Abstract

Banking supervisors worldwide recognise the pressing need to harness frontier technologies such as artificial intelligence (AI), particularly machine learning (ML), to enhance their efficiency and analytical capabilities. The European Central Bank (ECB) has similarly acknowledged the opportunities offered by supervisory technology (SupTech) and established a dedicated Suptech Hub. However, the adoption of automated technologies in banking supervision raises complex questions of legality, transparency, and accountability, particularly for the ECB, as a public institution within the EU’s democratic order founded on the rule of law.

This study investigates how the use of AI systems to augment supervisory decision-making may impact EU fundamental rights, particularly the right to good administration. To this end, we first define the notion of good administration in the context of banking supervision, and explore what it entails for the ECB from legal and ethical perspectives. We then analyse the potential implications of AI-enhanced banking supervision for good administration and examine how the latter may inform the integration of AI/ML into supervisory processes and procedures.

Drawing inspiration from the proposed EU AI Act, we develop a normative framework for regulating AI systems based on specific risks to good administration associated with different applications. Our framework prioritises transparency, auditability and accountability requirements to ensure that future AI-driven banking supervision is aligned with the principles of good administration. Overall, this study contributes to the growing literature on the legal implications of AI and ML adoption by financial supervisors, underscoring the importance of a balanced approach that upholds fundamental rights while harnessing the benefits of technological progress.

Keywords: artificial intelligence, machine learning, banking supervision, ECB, EU administrative law, good administration, fundamental rights, judicial review

JEL Classification: G18, G28, G38, K22, K42, O33, O38

Suggested Citation

Azzutti, Alessio and Batista, Pedro Magalhães and Ringe, Wolf-Georg, Good Administration in AI-Enhanced Banking Supervision: A Risk-Based Approach (April 27, 2023). European Banking Institute Working Paper Series 2023 - no. 140, Available at SSRN: https://ssrn.com/abstract=4430642 or http://dx.doi.org/10.2139/ssrn.4430642

Alessio Azzutti (Contact Author)

University of Glasgow - School of Law ( email )

Stair Building
5 - 8 The Square
Glasgow, Scotland G12 8QQ
United Kingdom

National University of Singapore (NUS) - Centre for Banking and Finance Law (CBFL) ( email )

Singapore

HOME PAGE: http://https://law.nus.edu.sg/cbfl/people/alessio-azzutti/

Institute of Law & Economics - University of Hamburg ( email )

Johnsallee 35
Hamburg, 20148
Germany

European Banking Institute ( email )

Frankfurt
Germany

HOME PAGE: http://ebi-europa.eu/

Pedro Magalhães Batista

University of Leeds ( email )

Leeds, LS3 1DB
United Kingdom

New York University School of Law ( email )

New York City, NY 10012
United States

Wolf-Georg Ringe

University of Hamburg - Institute of Law and Economics ( email )

Alsterterrasse 1
Hamburg, 20354
Germany
+49 40 42838 7787 (Phone)

HOME PAGE: http://www.ile-hamburg.de

University of Oxford - Faculty of Law

St Cross Building
St Cross Road
Oxford, OX1 3UL
United Kingdom

HOME PAGE: http://www.law.ox.ac.uk/people/wolf-georg-ringe

European Corporate Governance Institute (ECGI)

c/o the Royal Academies of Belgium
Rue Ducale 1 Hertogsstraat
1000 Brussels
Belgium

HOME PAGE: http://ecgi.global/users/wolf-georg-ringe

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
810
Abstract Views
1,958
Rank
61,971
PlumX Metrics