How AI can combat the growing menace of trade-based money laundering

By Colin Camp, senior director, business development & sales, APAC, Pelican

9 December 2019

The menace of trade-based money laundering (TBML) is an increasing, yet often under-reported, financial and reputational risk to banks and a growing concern to governments and regulators. Transnational crime is worth up to $2.2trn each year and much of it is facilitated by various forms of trade-based money laundering. A PWC report stated that 80 percent of illicit financial flows from developing countries are accomplished through trade-based money laundering.

With sums of this magnitude, it is not surprising that banks, who are often the (unwitting) facilitators of this illegal activity, are coming under increasing pressure from regulators to take greater action to limit this growing international crime. For banks, TBML, disguised under the huge volumes of legitimate trade, is also extremely difficult to detect. Techniques such as falsifying documents, under- or over-invoicing, and misrepresenting financial transactions, are difficult to trace as they involve multiple parties, jurisdictions and transactions and the information and data is usually in paper-based documents.

Regulatory focus

With this growing regulatory focus, banks are increasingly obliged to take more stringent action to ensure they are not facilitating any illicit transactions. Several international regulatory bodies have already issued detailed guidelines and red flag checks around trade finance, echoing those issued by the International Chamber of Commerce (ICC), Bankers Association for Finance and Trade (BAFT) and the Wolfsberg Group and more are expected to follow suit in 2020. These red flag checks define the key attributes in trade finance transactions that indicate a high risk for TBML and are now seen as the global standard for due diligence for which financial institutions must screen and monitor.

To keep track of trade finance transactions is a compliance headache. For many banks this is a manual process. For banks that have attempted automation, multiple applications and legacy systems are maintained for the different aspects of TBML compliance, such as sanctions, vessel tracking, and document matching. There is then a need to link these systems competently to a range of compliance filters, each requiring their own unique workflows. The paper-based processes prevalent along the trade-transaction chain make this compliance process, either manual or partially automated, extremely inefficient and often costly labour-intensive process.

TBML is difficult to detect because of the complexity within the trade finance process itself, the high number of entities and data sources involved, and a long-held reliance on paper-based documentation. A recent ICC study reported that that there are over four billion pages of documents currently circulating in trade at any one time.

The document challenge

The unstructured, inconsistent format of trade finance documentation, including text, PDF and image files, create automation and screening challenges. Trade instruments such as bills of exchange, bills of lading, letters of credit, invoices, insurance documentation and SWIFT MT7xx series are heavy on free-format text and unstructured data, and do not lend themselves well to most compliance filters, which require formatted, structured data in order to accurately and correctly detect non-compliant items.  In other words, it’s easy for a mispresented price or quantity of goods, or a false customs declaration to go unnoticed.

With such a complex task, with multiple document formats and data sources to monitor, a digitised process leveraging the unique capabilities of artificial intelligence (AI) technology providing augmented intelligence (helping humans become faster and more efficient on their tasks rather than completely replacing them) becomes the only solution for banks wishing to meet their TBML compliance obligations. Unstructured data from the various paper-based trade documents must first be scanned and put into machine-readable text format with the help of optical character recognition (OCR) technology. Once the data is in a format that can be processed and analysed, the AI discipline of natural language processing (NLP) can be used in combination with knowledge based techniques to interpret the text, understand the context and derive meaning from it and to extract key trade information automatically. AI-based tools are able to intelligently compare, identify and provide alerts for red flag indicators and sanctions subjects.

AI solutions

By harnessing the unique capabilities of AI technology, banks are able to go beyond the efficiency benefits of simple document management workflow, to be able to intelligently monitor and rapidly detect money laundering activities, without the burden of having to employ large numbers of expensive, error-prone and time-consuming human resources to tackle difficult compliance checks manually. By providing this augmented intelligence, trade finance operational and compliance staff can concentrate on real issues requiring investigative skills rather than the mundane manual tasks of reading and checking of documents and inputting of data into screening tools.

Transnational crime and TBML will continue to grow until the paradigm of high profits and low risks is challenged. However, with the practical application of AI in payments and financial crime compliance, and with end-to-end trade transaction cycle monitoring, banks are well placed to redress this balance and adapt to heightened regulation required to tackle this growing menace.

Become a bobsguide member to access the following

1. Unrestricted access to bobsguide
2. Send a proposal request
3. Insights delivered daily to your inbox
4. Career development