Artificial Intelligence, Algorithmic Herding and Systemic Fragility in Financial Markets

Saker Sabkha () and Hamdi Jbir ()
Additional contact information
Saker Sabkha: LEGO - Laboratoire d'Economie et de Gestion de l'Ouest - UBS - Université de Bretagne Sud - UBO EPE - Université de Brest - IMT - Institut Mines-Télécom [Paris] - IBSHS - Institut Brestois des Sciences de l'Homme et de la Société - UBO EPE - Université de Brest - UBL - Université Bretagne Loire - IMT Atlantique - IMT Atlantique - IMT - Institut Mines-Télécom [Paris]
Hamdi Jbir: LEGO - Laboratoire d'Economie et de Gestion de l'Ouest - UBS - Université de Bretagne Sud - UBO EPE - Université de Brest - IMT - Institut Mines-Télécom [Paris] - IBSHS - Institut Brestois des Sciences de l'Homme et de la Société - UBO EPE - Université de Brest - UBL - Université Bretagne Loire - IMT Atlantique - IMT Atlantique - IMT - Institut Mines-Télécom [Paris]

Working Papers from HAL

Abstract: This paper investigates whether the diffusion of artificial intelligence (AI) within financial institutions acts as a structural driver of herding behavior and amplifies systemic financial stress. While prior research highlights the micro-level efficiency gains from AI adoption, its collective and macro-financial implications remain largely unexplored. Using daily stock returns for major U.S. financial firms over 2010-2025, we measure herding through crosssectional return dispersion and construct a novel firm-level AI adoption index based on textual analysis of annual reports. This index captures structural AI intensity, adoption dynamics, and relative positioning within peer groups. Results show that herding is strongly state dependent, intensifying during market downturns. AI adoption does not increase herding on average; however, higher aggregate AI exposure significantly amplifies nonlinear herding under adverse market conditions, particularly after 2017. Sectoral analyses indicate that herding is more pronounced in market-facing and infrastructure-intensive financial activities, while AI-driven amplification primarily operates at the system level. At the same time, higher aggregate AI adoption is associated with lower baseline systemic stress, suggesting a trade-off between efficiency gains and synchronization risks. Overall, the findings identify AI as a coordination technology that enhances efficiency in normal times but can amplify collective dynamics during stress, with important implications for financial stability monitoring.

Keywords: Artificial intelligence Algorithmic herding Systemic risk Financial stability Market synchronization; Artificial intelligence; Algorithmic herding; Systemic risk; Financial stability; Market synchronization (search for similar items in EconPapers)
Date: 2025-10-21
Note: View the original document on HAL open archive server: https://hal.science/hal-05556083v1
References: Add references at CitEc
Citations:

Downloads: (external link)
https://hal.science/hal-05556083v1/document (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:hal:wpaper:hal-05556083

Access Statistics for this paper

More papers in Working Papers from HAL
Bibliographic data for series maintained by CCSD ().