Trade-Based Money Laundering

How Technology Is Changing the Fight Against One of the Biggest Financial Crimes

by Andrea Frosinini and Viktoria Soltesz

Trade-based money laundering (TBML) is one of the most significant and difficult-to-detect threats in global financial crime where criminals use the complexity of international trade to move illicit funds on a large scale. Every day, trillions of dollars in goods cross borders through complex documents and multiple jurisdictions, giving criminals an effective cover. The exact scale of TBML is hard to measure, but the FATF and the World Economic Forum estimate it generates around USD 1.6 trillion in illicit flows each year: in developing markets, up to 80% of illicit financial flows are believed to move through trade channels.

A recent analysis discussed at the FATF/GIABA joint meeting in Ghana showed that commercial banks there facilitated roughly US$20 billion in foreign transfers with no matching imports between April 2020 and August 2025. This highlights major gaps in monitoring and serious capital flight risks. This report explains how TBML works, why it is so hard to detect, and how new technologies including artificial intelligence, machine learning, blockchain, and advanced analytics are now being used to fight it. Advanced technology has become essential for protecting the integrity of the global financial system.

Understanding Trade-Based Money Laundering

Definition and Mechanisms

TBML is the process of disguising the proceeds of crime through international trade transactions. It is one of three main methods used by criminal networks, alongside cash smuggling and the formal banking system.

The FATF defines TBML as disguising criminal proceeds and moving value through trade to make them appear legitimate. Common techniques include:

• Over- or under-invoicing of goods and services

• Multiple invoicing of the same shipment

• Falsely describing goods

• Creating phantom shipments for goods that never exist

• Shipping different quantities or qualities than stated on documents

The 2025 FATF risks review noted a rise in multi-layered shipments and ship-to-ship transfers that make detection even more difficult.

The Trade Finance Nexus

TBML exploits trade finance instruments such as letters of credit, bank guarantees, and bills of lading. These tools were designed for legitimate commerce long before large-scale financial crime became common. A 2025 academic study showed that the huge volume of trade documents makes manual review almost impossible for banks.

Multiple parties are involved in every trade (exporters, importers, freight forwarders, customs, banks, and insurers), but no single party sees the full picture. This fragmentation is what criminals exploit.

The Regulatory Landscape

Regulators have tightened rules in recent years. 

The FATF 2024–2025 Annual Report, published in March 2026, updated global standards and highlighted how criminals combine trade methods with shell companies and virtual assets to evade sanctions. The November 2025 FATF/GIABA joint experts meeting in Accra placed trade-based crime at the top of the agenda. The UK's HMRC has also emphasised the need for advanced technology, data analytics, and international cooperation because traditional methods are no longer sufficient.

Why TBML Is So Hard to Detect

Banks only see the payment side of a transaction, without visibility on what goods are actually being shipped, while customs authorities focus on the physical goods but do not have access to the financial details behind the transaction. This separation of information creates gaps, and a 2025 analysis confirmed that these gaps are one of the main reasons why proper detection is difficult.

International trade produces a very large number of documents, including invoices, bills of lading, certificates of origin, and SWIFT messages, all of which are detailed and require careful review. Handling this volume manually is simply not realistic, especially when transactions increase in scale and speed.

Another issue comes from the difference in how fast different players are advancing. Criminal networks are already using AI and automation to improve their methods, while many banks still rely on manual processes and older systems. A March 2026 industry report highlighted that financial institutions are falling behind because of this gap.

Tariff evasion is also becoming a growing concern, as it often uses the same techniques as trade-based money laundering. A 2025 analysis showed that these two areas are closely linked, which means that combining anti-money laundering checks with trade compliance has become much more important, especially as global trade tensions continue to increase.

The Technology Arsenal: Current Tools

Financial institutions have spent years chasing TBML with tools built for a different era, keyword searches, rule-based alerts, and manual document reviews that simply cannot keep pace with the volume or sophistication of modern trade crime. That is beginning to change, and in some areas the results are measurable.

Machine learning is where the most tangible progress is being made. A March 2026 paper in MDPI's Information journal introduced a model called BO-XGBoost, which addresses one of the biggest practical frustrations in AML work: the flood of false positives that consumes analyst time without producing results. By handling uneven datasets more effectively, BO-XGBoost keeps detection rates high while cutting unnecessary alerts. Another model, FALCON, published in August 2025 in the Journal of Risk and Financial Management, takes a different approach, combining multiple AI techniques to surface patterns that no single method would catch alone. Both represent a meaningful shift: from systems that flag anomalies to ones that begin to understand context.

Document analysis is also improving in ways that matter operationally. Systems using retrieval-augmented generation (RAG) can now read trade documents for meaning rather than simply scanning for keywords. In practice, this means spotting inconsistencies between an invoice, a bill of lading, and a customs declaration that would pass a keyword search, but fall apart when the documents are read together.

Blockchain offers something structurally different: a way to make the trade document record itself harder to manipulate. A May 2025 SSRN paper showed how distributed ledger technology can create permanent, tamper-resistant records across the multiple parties involved in a single shipment. The promise is not just greater transparency but also removing the window in which documents can be quietly altered between origination and settlement. 

One less-discussed but significant development is synthetic data. Because real TBML case data is rare, sensitive, and unevenly distributed, training AI models on it has always been difficult. A May 2025 Springer paper introduced a framework for generating realistic artificial datasets that mirror the structure of genuine TBML activity, giving models far more material to learn from without the privacy and legal complications of working with actual case files.

However, even the best tools face real-world difficulties: high false positive rates, lack of explainability for regulators, strict data privacy rules that limit sharing, and uneven adoption between large and small institutions.

The Road Ahead: Strategic Imperatives and Planning

Success requires a shift from detection to prevention, better public-private data collaboration, convergence of AML and trade compliance, and continuous preparation for an AI arms race with criminals.

What often receives less attention in these discussions is how payment and banking flows are structured in the first place. TBML does not only succeed because detection fails; it also succeeds because the underlying payment architecture leaves room to operate. How transactions are structured, how funds move between counterparties, which banks and payment providers are selected, and how documentation is aligned across parties all create or close the gaps that criminal networks exploit. These are not merely operational decisions; they are strategic ones, and they carry direct compliance consequences.

The challenge is that in most organisations, no single function owns them. Payment and banking decisions tend to be distributed across finance, compliance, operations, and product teams, each managing a piece of the picture without full visibility into the whole. Controls developed in isolation rarely account for what happens at the handoffs between teams, and that is precisely where exposure tends to accumulate.

This also highlights the importance of ownership. In many organisations, payment and banking decisions are spread across finance, compliance, operations, and product teams, with no single point of responsibility. This lack of coordination makes effective prevention more difficult.

That structural gap points toward a structural answer.

The role of a Chief Payment Officer addresses this gap. By taking ownership of payment flows, banking relationships, and the overall structure of how money moves, this role ensures that systems, controls, and processes are aligned from the start. This creates a stronger foundation for both compliance and operational stability in international trade.

Conclusion

Trade-based money laundering remains a major threat to the global financial system, estimated at USD 1.6 trillion annually. Technology now offers powerful ways to detect and prevent it, but tools alone are not enough. Real progress depends on better data sharing, clearer regulations, and stronger institutional commitment.

Why Structured Payment and Banking Planning Is Absolutely Essential

Payment and banking flows sit at the centre of every trade transaction and are often the first place where TBML risks appear. In today's complex environment, these flows must be treated as a standalone strategic function, not just a part of finance.

This is where the Chief Payment Officer becomes essential. A dedicated CPayO provides clear ownership and oversight of the entire money flow, turning fragmented efforts into one coherent strategy. For companies involved in international trade, structured payment and banking planning combined with CPayO leadership protects cash flow, reduces blind risks, strengthens compliance, and builds the resilience needed to operate securely when criminal tactics continue to evolve.

by Andrea Frosinini and Viktoria Soltesz