美联储-生成型人工智能对金融稳定的影响：驯服动物精神（英）

Finance and Economics Discussion SeriesFederal Reserve Board, Washington, D.C.ISSN 1936-2854 (Print)ISSN 2767-3898 (Online)Financial Stability Implications of Generative AI: Taming theAnimal SpiritsAnne Lundgaard Hansen, Seung Jung Lee2025-090Please cite this paper as:Hansen, Anne L., and Seung Jung Lee (2025).“Financial Stability Implications ofGenerative AI: Taming the Animal Spirits,” Finance and Economics Discussion Se-ries 2025-090.Washington:Board of Governors of the Federal Reserve System,https://doi.org/10.17016/FEDS.2025.090.NOTE: Staff working papers in the Finance and Economics Discussion Series (FEDS) are preliminarymaterials circulated to stimulate discussion and critical comment. The analysis and conclusions set forthare those of the authors and do not indicate concurrence by other members of the research staff or theBoard of Governors. References in publications to the Finance and Economics Discussion Series (other thanacknowledgement) should be cleared with the author(s) to protect the tentative character of these papers.Financial Stability Implications ofGenerative AI: Taming the Animal Spirits ∗Anne Lundgaard Hansena,b and Seung Jung LeeaaBoard of Governors of the Federal Reserve SystembFederal Reserve Bank of RichmondSeptember 25, 2025This paper investigates the impact of the adoption of generative AI on financial stability. We conductlaboratory-style experiments using large language models to replicate classic studies on herd behaviorin investment decisions. Our results show that AI agents make more rational decisions than humans,relying predominantly on private information over market trends. Increased reliance on AI-poweredinvestment advice could therefore potentially lead to fewer asset price bubbles arising from animal spiritsthat trade by following the herd. However, exploring variations in the experimental settings reveals thatAI agents can be induced to herd optimally when explicitly guided to make profit-maximizing decisions.While optimal herding improves market discipline, this behavior still carries potential implications forfinancial stability. In other experimental variations, we show that AI agents are not purely algorithmic,but have inherited some elements of human conditioning and bias.Keywords: Herd behavior, large language models, AI-powered traders, financial markets, financial sta-bility.JEL Codes: C90, D82, G11, G14, G40.∗ The authors thank Jeffrey S. Allen, Marco Cipriani, Erik Heitfield, Sophia Kazinnik, Dan Li, Molly Mahar, andNitish Sinha and participants at the Financial Stability Workshop at the Federal Reserve Board for valuable feed-back. The views expressed in this paper are solely those of the authors and do not reflect the opinions of theFederal Reserve Bank of Richmond or the Board of Governors of the Federal Reserve System. Generative AI wasused in the production of this paper. All errors are the authors’. Contact information: anne.hansen@rich.frb.org,seung.j.lee@frb.gov.1...[T]here is the instab