An Agentic LLM Framework for Adverse Media Screening in AML Compliance

TL;DR

This paper introduces an agentic LLM-based system that automates adverse media screening for AML compliance by retrieving, processing, and scoring relevant information to identify high-risk individuals more efficiently.

Contribution

It presents a novel multi-step LLM framework utilizing Retrieval-Augmented Generation to improve accuracy and automation in adverse media screening for AML.

Findings

The system effectively distinguishes high-risk from low-risk individuals.

It demonstrates strong performance across diverse datasets including PEPs and sanctions.

The approach reduces false positives compared to traditional keyword-based methods.

Abstract

Adverse media screening is a critical component of anti-money laundering (AML) and know-your-customer (KYC) compliance processes in financial institutions. Traditional approaches rely on keyword-based searches that generate high false-positive rates or require extensive manual review. We present an agentic system that leverages Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to automate adverse media screening. Our system implements a multi-step approach where an LLM agent searches the web, retrieves and processes relevant documents, and computes an Adverse Media Index (AMI) score for each subject. We evaluate our approach using multiple LLM backends on a dataset comprising Politically Exposed Persons (PEPs), persons from regulatory watchlists, and sanctioned persons from OpenSanctions and clean names from academic sources, demonstrating the system's ability to distinguish between high-risk and low-risk individuals.