Agentic AI helps trace real illicit flows on Ethereum blockchain
AMLBench: A Real-Incident Benchmark for Agentic Blockchain Anti-Money Laundering
Computational Engineering, Finance, and Science
Summary
Tracing illegal money on public blockchains like Ethereum is hard because funds move through many accounts and get mixed up. The authors created AMLBench, a benchmark from 24 real cases and millions of transactions, where AI agents try to follow the money starting with little information. They also built AMLAgent, an AI tool that uses smart guidance to decide which accounts to check. This AI traced money with over 80% accuracy, showing promise for more reliable and faster automated money laundering investigations.
What this means in practice
- •For blockchain compliance teams: Automate tracing suspicious transactions on Ethereum using AI agents guided by transaction queries and fund flow tools.
- •For cybersecurity analysts: Use an AI benchmark and tracing agent to improve investigation efficiency and insight into illicit fund flows.
Authors
Qishuang Fu, Hang Zheng, Yuchen Lei, Yuexin Xiang, Qin Wang, Yuanzhe Zhang, Shiping Chen, Andreas Deppeler, Joseph Liu, Ziyao Liu, Tsz Hon Yuen
Abstract
Tracing illicit funds on public blockchains remains largely manual. As funds move across accounts and undergo asset conversions, they become mixed with other inflows, requiring repeated queries and expert judgment to trace. Large language model (LLM) agents offer a way to automate this process, but their ability to reconstruct case-related transfer networks from limited clues remains unclear. To bridge this gap, we present AMLBench, an anti-money laundering (AML) benchmark built from 24 real-world cases and 5.05 million Ethereum transfer records. For each case, agents start from limited clues and explore these transfer records through composable transaction-query and fund-flow analysis tools, deciding which accounts and branches to pursue. The final tracing results are evaluated against case annotations, while tool interaction logs reveal how agents discover and select relevant transfers and at what resource cost. We also develop AMLAgent, which integrates fund-tracing guidance and query feedback to dynamically guide investigative decisions. With GPT-6 Astra, AMLAgent achieves macro averages of 80.82% transfer-level precision and 84.73% query coverage across 24 real-world cases. Further analysis of the large-scale Bybit case reveals substantial differences in transfer retention and token usage across models, even with similar submission precision. AMLBench establishes a reproducible benchmark for evaluating both tracing outcomes and investigative processes, laying the groundwork for more reliable and efficient autonomous AML investigations.