An Agentic LLM Framework for Adverse Media Screening in AML Compliance

Adverse media screening is a critical component of anti-money laundering (AML) compliance in financial institutions. Traditional keyword-based approaches generate high false-positive rates, while hybrid systems require extensive manual review. We present an agentic system leveraging Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to automate adverse media screening. Our system implements a multi-step pipeline where an LLM agent searches the web, retrieves and processes relevant documents, and computes an Adverse Media Index (AMI) score with natural language justifications. The open-source implementation supports multiple LLM backends including local and API-based services.