Generative AI and Agentic AI in Financial Crimes Compliance

Abstract Generative Artificial Intelligence (“AI”) (GenAI) (e.g. large language models) and agentic AI (autonomous, goal directed AI agents) offer powerful new capabilities for financial-crime compliance, including Anti Money Laundering (“AML”)/Combating Finance of Terrorism (“CFT”) programs, customer due diligence, transaction monitoring, and investigations. These technologies can automate and scale tasks such as Suspicious Activity Report(“SAR”) narrative drafting, alert triage, and data analysis, potentially improving detection rates and freeing human analysts for complex work. However, they introduce novel risks - for example, “confabulation” (AI-generated false or misleading content), privacy and bias issues, and loss of human control over automated processes. Regulatory guidance is only now beginning to address these innovations: U.S. bank regulators have explicitly excluded generative/agentic AI from standard model risk guidance and signaled dedicated future oversight. In this context, financial institutions must treat generative/agentic AI as high-risk, applying rigorous governance, explainability, data controls, and human-in-the-loop safeguards. Key recommendations include: pilot projects with regulators, robust validation against factual accuracy, clear audit trails, and continuous monitoring for model drift and misuse. This white paper reviews the state of technology and regulation, identifies technical and control challenges, and offers actionable guidance for compliance leaders and model-risk teams.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24323894
Primary Topic
Ethics and Social Impacts of AI
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Generative AI and Agentic AI in Financial Crimes Compliance

Parth Trivedi
2 links
Zenodo (CERN European Organization for Nuclear Research)
Ethics and Social Impacts of AI
article

Generative AI and Agentic AI in Financial Crimes Compliance

Parth Trivedi
article en

Abstract

Abstract Generative Artificial Intelligence (“AI”) (GenAI) (e.g. large language models) and agentic AI (autonomous, goal directed AI agents) offer powerful new capabilities for financial-crime compliance, including Anti Money Laundering (“AML”)/Combating Finance of Terrorism (“CFT”) programs, customer due diligence, transaction monitoring, and investigations. These technologies can automate and scale tasks such as Suspicious Activity Report(“SAR”) narrative drafting, alert triage, and data analysis, potentially improving detection rates and freeing human analysts for complex work. However, they introduce novel risks - for example, “confabulation” (AI-generated false or misleading content), privacy and bias issues, and loss of human control over automated processes. Regulatory guidance is only now beginning to address these innovations: U.S. bank regulators have explicitly excluded generative/agentic AI from standard model risk guidance and signaled dedicated future oversight. In this context, financial institutions must treat generative/agentic AI as high-risk, applying rigorous governance, explainability, data controls, and human-in-the-loop safeguards. Key recommendations include: pilot projects with regulators, robust validation against factual accuracy, clear audit trails, and continuous monitoring for model drift and misuse. This white paper reviews the state of technology and regulation, identifies technical and control challenges, and offers actionable guidance for compliance leaders and model-risk teams.

Zenodo (CERN European Organization for Nuclear Research)
Peace, justice, and strong institutions
Ethics and Social Impacts of AI
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon