All articles

Harnessing AI for next-level AML

Transaction monitoring against money laundering (AML), and other categories of fraud line activity, including terrorism financing (CTF), constitutes a crucial challenge, both for financial institutions with funds and reputations at stake and for solution providers like Applica, whose resources are much more progressive than the norms in the relevant sector.

One reason for this is administrative and cultural. Monetary authorities – including the Federal Reserve and the vast majority of its foreign counterparts – skew conservative about allowing new technologies to enter into the compliance process, particularly solutions endowed with “a mind of their own,” such as those based on AI. What is permitted by law is familiar, limited and stringently regulated. Thus, the possibilities for innovation lag behind those allowed in other areas of business and finance.

Another reason this is so challenging is technical and has to do with the stark disproportion in volume between the number of non-fraudulent and fraudulent transactions. It is a great irony within this critical slice of finance that the very activity that might result in billion-dollar losses to national and global economies – and baseline-busting legal harm to a negligent bank – constitutes such a minor fraction of day-to-day operations at any given financial institution. Thus, a 99.9% perfect know-your-client (KYC) protocol can in fact be viewed as blind to a vast one-tenth of one percent of customers and their transactions. And these may of course be hiding any number of fraudsters and important units of evidence.

In an increasingly cash-wary world, money laundering on an international scale is now a lifeline for practically all drug cartel operators, terrorist organizations, human trafficking rings, arms dealers, art thieves, and sellers of conflict diamonds. Additionally, it involves others, whose criminality is less like an action movie plot and much more simply about cooking the books. In most cases, a matryoshka-doll structure of shell companies is involved, as well as telltale – though often obscure and complex – patterns of multiple small transactions. Fraudsters move money many times between sub-companies and they move amounts just under the threshold figure flagged by an institution for mandatory review. Though much of the time detailed analysis at the client onboarding phase can alert a bank’s risk analysis team to potential fraud line activity, needles do get lost in haystacks – even with the best people in play.

Because robust compliance protocols are reliant not just on the experts in charge of the final assessment, but firsthand on the processing of high volumes of data, AI can and does help eliminate error and minimize risk. This is especially true in the case of business clients, as opposed to individual customers, because of the greater number of documents and resulting data points required at the onboarding phase. And when there is more content to verify, verification can run deeper. Legal entity name changes. Scanning for multiple accounts and suspicious financial records attached to names and IDs. Ongoing watchlist matching. Regional sanctions alerts. Examining sources of funds. Cross-comparison among the documents provided, for errors of fact and suspicious omissions. This is time-intensive work that can wear out risk analysts before they even get to the real analyzing. It’s a beautiful thing that AI can at last step in and assist – even if it is still prevented by law from taking over the process altogether.

Applica machine learning is highly customizable and compatible with extreme volumetric differences in data set size, including those resulting from the disproportionate amount of non-fraudulent activity in the financial sector. Thanks to this level of adaptability, the scrutiny, comparison, and sorting necessary for optimized AML, CTF, and KYC workflows can now be automated and performed more rapidly, accurately, and strategically. The costs this helps reduce extend beyond a financial institution’s bottom line and include essential “savings” in reputational, legal, and workforce terms. From speeding document review and pre-empting false positives to increasing safety rates and optimizing the productivity of an institution’s top compliance team analysts, Applica can make a vital contribution to a financial institution’s AML and CTF security and client onboarding success.

Interested in learning how Applica Robotic Text Automation can help solve your AML challenges?

Book a demo with an Applica expert today.