As part of their regulatory compliance obligations, banks and other financial institutions must develop and implement risk-based AML programs to counter the money laundering and terrorism financing threats that they face. After performing risk assessments on their customers, firms should ensure that suitable monitoring and screening measures are in place to alert them when those customers engage in activity that could be indicative of money laundering.
However, an AML alert does not necessarily mean that money laundering is taking place. The sensitivity of transaction monitoring and screening measures often means that firms have to deal with a significant amount of false positive AML alerts in their efforts to detect genuine criminal activity. In fact, false positives constitute a significant proportion of the alerts that AML monitoring and screening measures generate: some estimates even suggest that false positives account for around 42% of AML alerts and cost firms over $3 billion every year.
What Are False Positives?
The parameters of transaction monitoring and screening measures may make certain innocent customer behaviors appear suspicious. A customer that makes several cash withdrawals from different bank branches on the same day, for example, may trigger an AML alert even though they have legitimate business reasons for their transactional behavior. Similarly, certain Arabic naming conventions can result in different customers having very similar names and cause sanctions screening measures to incorrectly connect them to names on sanctions lists.
In both instances, the false positive AML alerts would need to be scrutinized by the financial institutions’ AML compliance teams, which may need to freeze the accounts of customers involved during their remediation process. In some cases, firms may not be able to resolve an AML alert as a false positive and so move the erroneous alert up the chain to a financial authority.
False positives represent a compliance challenge for financial institutions in jurisdictions around the world. However, while false positive alerts are significant cost and efficiency drains, firms often cannot afford to reduce the sensitivity of their monitoring and screening measures because of the potential financial, reputational and even criminal penalties that they face if they fail to meet their regulatory obligations. False positives also create negative customer experiences and, given the administrative work they generate, hinder firms’ efforts to address genuine money laundering incidents in the global fight against financial crime.
Finding a way to reduce false positives without compromising the accuracy and effectiveness of AML transaction monitoring and screening should be a priority for every compliance team. Accordingly, firms should understand how false positives occur and what strategies may help to reduce false positive rates.
While it may not be possible to eliminate them completely, a number of approaches may be effective in helping firms reduce their false positive rates:
Data structuring: Transaction monitoring and screening measures require firms to process and analyze a vast amount of data from a variety of sources, often in an unstructured format. The more confusing the data collected as part of procedural compliance, the more challenging it is to discern false positives from true positive AML alerts.
By considering the data capture process more carefully, and structuring that data clearly once acquired, firms may be able to improve their false positive rates. Practically, this means organizing names, for example, into title, first name and surname rather than listing each name as a single data point — an approach that is liable to create administrative ambiguity and hinder compliance teams’ attempts to resolve customer identities after AML alerts.
Data relevance: Criminals may seek to change their names or move between countries as a way to thwart AML controls. With that in mind, while the quantity of AML data captured is important to building an accurate profile of a customer, the relevance of that data is crucial to the verification process.
Firms should ensure that the data they collect on their customers serves the AML process by being relevant and timely to the customer’s risk profile. A false positive might be created, for example, if a firm fails to update a customer’s change of name or change of residence from a high-risk jurisdiction to a low-risk jurisdiction
Ongoing review: AML compliance should not be an exercise in checking boxes. Instead AML programs should grow and adapt to their environment as new criminal methodologies emerge or as new regulations are introduced. Accordingly, firms should conduct ongoing reviews of their screening and monitoring measures to ensure their continued accuracy and effectiveness.
With that in mind, it may be possible to adjust or remove certain AML controls depending on the regulatory environment, and so reduce the false positive rates in a safe and controlled manner. Similarly, certain money laundering methodologies may become obsolete thanks to technological advances, further reducing the AML compliance burden and the type of alerts that out-of-date detection measures generate.
Smart technology: Smart technology is a crucial component of the modern financial compliance process. But with the benefit of integrated artificial intelligence (AI) and machine learning models, firms may be able to add even greater efficiency and accuracy to their AML response, building richer, more relevant repositories of data for use in risk assessment.
In helping to reduce false positives, AI algorithms allow firms to analyze AML alerts faster and more accurately than human compliance teams. In practice, this means firms can streamline the alert remediation process by identifying and prioritizing certain cases and bringing them to the attention of compliance officers for review.
Machine learning models complement the speed and efficiency benefits of AI by allowing AML systems to exploit previously collected data in order to respond more effectively to new and future alerts. Practically, machine learning may enable firms to perform semantic and statistical analyses on new alerts, quickly identifying duplicate or redundant data that might be creating the appearance of suspicious behavior. Similarly, intuitive screening processes, based on machine learning models may be able to detect confusing customer naming conventions that are generating false positive hits on sanctions lists.
While AI and machine learning models shouldn’t be seen as a replacement for human expertise within the AML process, they can help compliance teams address false positive alerts in a much more structured and efficient manner. Accordingly, optimizing the integration of AI and machine learning should become a foundational part of the AML process and an important step towards reducing a firm’s false positive hit rate.
Download Customer Screening Report
This report will cover how you can find the right balance between customer screening effectiveness and efficiency, and how technology can help you.