Artificial Intelligence and Bank Intermediation: A Critical Review of the Evidence

Tudor Dragan, Simona Nistor and Steven Ongena

No 22129, CEPR Discussion Papers from Centre for Economic Policy Research

Abstract: We critically review the evidence on how artificial intelligence (AI) affects bank intermediation, from information production and lending decisions to bank performance and financial stability. Studies reach conflicting conclusions on profitability, risk and fragility, mainly because they measure AI, select outcomes and identify effects differently. We develop a framework that traces AI effects along the intermediation chain, from prediction to lending, bank balance sheets and the banking system, and explains why predictive gains pass through only partly and may change sign. Appraising the core studies on their AI measures, outcome–mechanism alignment and identification, we find that machine learning improves default prediction, but evidence that AI raises bank profitability or reduces bank risk awaits studies that combine direct AI measures with causal identification. We derive testable propositions on decision-chain attenuation, the diffusion of AI rents, within-bank risk reallocation, digital bank runs, credit-cycle amplification and generative AI, each with a research design.

Keywords: Artificial; intelligence (search for similar items in EconPapers)
JEL-codes: C55 (search for similar items in EconPapers)
Date: 2026-11
References: Add references at CitEc
Citations:

Downloads: (external link)
https://cepr.org/publications/DP22129 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:cpr:ceprdp:22129

Ordering information: This working paper can be ordered from
https://cepr.org/publications/DP22129

Access Statistics for this paper

More papers in CEPR Discussion Papers from Centre for Economic Policy Research 33 Great Sutton Street, London EC1V 0DX, UK.
Bibliographic data for series maintained by CEPR ().