How AI-Infused Approaches to AML Compliance Are Changing the Game

By Greg Pinn

Advances in data analytics and artificial Intelligence (AI) are giving organizations powerful capabilities to reimagine existing operational models and approaches to risk management, including anti-money laundering (AML) compliance efforts.

Manual, human remediation of screening matches is the largest area of cost associated with AML compliance. A recent study found that 40% of financial crime compliance costs are related to internal labor costs with an additional 20% for outsourced costs. With an average staff tenure of 12 months, financial institutions must continually hire and train new AML analysts to cover both attrition and growing compliance needs. These costs continue to grow both in absolute terms and as a percentage of overall compliance spend.

Moreover, traditional AML compliance approaches are ineffective in keeping up with evolving threat vectors. Banks and other financial institutions were fined almost $5 billion for AML infractions in 2022.

To address these challenges, AI and data analytics are being used to more efficiently and effectively combat money laundering, tamp down compliance costs, and avoid infractions. AI-infused approaches are helping compliance teams supplement existing systems and reduce manual remediation efforts for watchlist screening. These new approaches also can provide accuracy scores as well as explanations for how they arrived at assessments, which is important in satisfying regulators. (However, it should be noted that regulators have taken a very positive approach to incorporating new technologies in the combat against money laundering.)

Many financial institutions are deeply invested in name matching systems that can’t be easily replaced, leading to extensive manual effort in sifting through extensive false positives to find the true positive.  While replacing these existing systems would often lead to improved operational efficiency and reduced human error, adding post-processing to existing systems can be an effective remedy to this situation.

By incorporating an application programming interface (API) over businesses’ existing customer identification systems (whether on-site or in the cloud) or AML matching engine, the latest AI-enabled approaches to AML compliance can help organizations avoid the need to replace expensive systems to improve client and third-party onboarding processes. This is done at faster-than-human speeds and with greater accuracy, which cuts labor costs by substantially reducing the requirement for human intervention. Plus, regular system updates keep the organization abreast of the latest regulatory obligations and the shifting threat landscape.

Infusing the power of AI into AML compliance boosts critical capabilities. Using sophisticated natural language processing (NLP) capabilities, AI-powered name matching systems compare names to sanctions, watch lists, and custom lists; and incorporate a broad variety of languages, detecting aliases, nicknames, and misspellings. Improved accuracy comes from recognition of naming variations across cultures by having the ability to incorporate nearly two dozen languages, regardless of their script.

The best name matching technologies also provide clear scores and details on match phenomena, helping users to understand why names have been deemed “matches” or “mismatches,” instilling confidence in the results obtained. These systems also empower users to adjust match parameters according to an organization’s data profile needs and risk policies.

When incorporated with an existing system, advanced AI-infused AML systems post-process the output from the existing system system to add accuracy for business-critical use cases.

The process of finding the right people is accomplished in part by matching entities to other identifiers, including dates of birth, places of birth, relatives, social media handles, and other identification records.

Advanced AML technology helps reduce the overall cost of compliance and improves compliance reporting by ensuring a seamless customer experience via faster, easier onboarding. Fast and accurate name matching improves legitimate customer experiences by minimizing payment delays, rejections, or disputes that could otherwise arise due to name matching discrepancies.

Some advanced AML name matching solutions can indicate match confidence scores and provide explanations for how they arrived at the assessment, to further satisfy regulatory reporting requirements and provide for advanced system tuning.

By leveraging the advanced natural language processing (NLP) capabilities on which anti-money laundering (AML) technology is built and having the ability to correctly filter extensive data sources, financial institutions can change the rules of the game to their favor and reduce errors as well as risk. Advanced AML technology can uncover intricate relationships and potential bad actors that may elude traditional screening processes, while protecting or even enhancing experiences for honest, legitimate customers. At the same time, AML systems can help ensure regulatory compliance, which reduces the risk of costly fines and penalties, while also eliminating some labor needs and these related costs.

About Greg Pinn

Greg Pinn is Babel Street’s Director of Name Screening in charge of anti-money laundering (AML), know your customer (KYC), and customer due diligence (CDD) product development. He has extensive experience in regulatory compliance with a focus on bringing products and solutions to market to mitigate regulatory and reputational risk faced by financial institutions and regulated businesses. Greg holds a Bachelor of Science in Computer Engineering from the Georgia Institute of Technology.