Navigating the Labyrinth: Artificial Intelligence in the Battle Against Trade-Based Money Laundering

An audio summary of this article is available in the player below. Scroll to keep reading. 
Listen and subscribe to Womble Perspectives wherever you get your podcasts.

In the rapidly evolving financial landscape, governments worldwide are intensifying their focus on compliance in Know Your Customer (KYC) and Anti-Money Laundering (AML) efforts for financial institutions. Trade-Based Money Laundering (TBML), a particularly challenging form of money laundering, has prompted increased oversight and regulation. This article explores the historical difficulties in detecting TBML, emphasizes its magnitude, and delves into the role of artificial intelligence (AI) in effectively combating this pervasive issue. Additionally, we provide an update on recent legislative developments aimed at addressing cross-border financial crimes, including TBML.

TBML involves disguising illicit funds through legitimate trade transactions, exploiting the complexity of the global trade system. Criminals employ techniques such as over-invoicing, under-invoicing, and misrepresentation of goods to move money across borders undetected, concealing it within legitimate business activities.

TBML is a global issue, potentially accounting for up to 80% of illicit financial flows globally, according to a report by Global Financial Integrity. This alarming scale underscores the need to address TBML effectively, as it undermines economic stability, facilitates organized crime, and finances terrorism.

Detecting TBML poses significant challenges due to the complexity and vast volume of global trade transactions. The lack of harmonized data and information sharing among multiple parties across different jurisdictions hampers the identification of suspicious patterns. Moreover, the sophistication of money laundering techniques, including transshipment, falsified trade documents, and manipulated quantity, quality, or price of goods, complicates traditional detection methods. Comprehensive screening is critical to identifying TBML, but comes with the challenge of false positives, which can have a chilling economic effect.  Unnecessarily putting certain technologies, providers, or components on a sanctions list, either with mistaken name matching, or overzealous policy, can put legitimate companies at a disadvantage. Therefore, the vigilance of screening must be balanced with the demand for selectivity and accuracy.

Financial institutions have relied on a tedious, manual review of the alerts produced from such screening platforms. The majority of these solutions include some rudimentary machine learning techniques, but very few are taking advantage of emerging, advanced AI that enables efficiency and more closely emulates human logic for problem solving. These new tools evaluate large volumes of complex data and do more than generate rules-based alerts on unusual transactions. They are assessing risk after a holistic analysis of the input data.

This next generation of AI solutions are imperative for effective compliance programs, and produce far fewer false positives than previous tools while augmenting investigators in handling the sheer volume of bad actors and crimes associated with TBML.

AI-backed software can be utilized to screen all trade parties, across all dimensions of internal and external data. The data attributes analyzed identify activity profiles that may indicate risk signals related to wildlife trafficking, drug trafficking, fraud, bribery, corruption, organized crime, tax evasion, and money laundering. Additionally, AI tools can swiftly extract relationships from structured and unstructured data feeds, helping investigators build a comprehensive network of business relationships and beneficial owners associated with any target entity.

The Combating Cross-border Financial Crime Act of 2023 was introduced by U.S. Senators Bill Cassidy, Sheldon Whitehouse, and Angus King. This legislation aims to establish a Cross-Border Financial Crime Center within the Department of Homeland Security, focusing on coordinating investigations and information sharing related to financial crimes, including TBML, with a nexus to the U.S. border. The proposed Center would serve as a central hub, housed within the lead criminal investigation arm of the Department of Homeland Security, Homeland Security Investigations, to analyze and coordinate financial crime data and investigations across the federal government.

As governments and regulators become more informed about the capabilities of innovative AI-based technologies, financial institutions must embrace these solutions to effectively combat TBML. The legislative developments highlight the ongoing commitment to addressing cross-border financial crimes and fortifying the regulatory framework in the fight against TBML. Rigorous compliance programs should integrate AI components and automation, evolving with technology to stay ahead of emerging threats in the financial landscape.

Howard W. Herndon is a Partner with Womble Bond Dickinson (US) LLP in the firm’s Fintech Practice.  He focuses his practice on the electronic transaction industry. For over two decades, he has represented public and private payments companies in significant industry transactions ranging from US $100 million to over US $1 billion. He is also a Managing Director and Founder of Prescentus, a subsidiary of Womble Bond Dickinson (US) LLP that offers full-service strategic business guidance for Fintech companies.

Robert “Al” Broadbent represents clients in international trade and national security matters and government investigations in the commercial and defense sectors. He also serves as a Managing Director of Prescentus. His practice includes guiding clients, including financial services companies, in such areas of international trade and national security as export controls, economic/trade sanctions, foreign direct investment/national security reviews and investigations (CFIUS), and more.

Chrissy Park is a Senior Director of Product at Quantifind. Quantifind was founded in 2009 upon pioneering work building machine learning technology to discover meaningful patterns across large, disparate, unstructured datasets. Quantifind’s Graphyte platform is differentiated by its risk assessment accuracy and speed, achieved through best-in-class name science, AI-driven entity resolution, dynamic risk typologies, real-time relationship extraction, and patented in-memory data storage and search techniques. The platform embodies over a decade of R&D and large-scale deployments with government agencies and Fortune 50 companies.