Covering Disruptive Technology Powering Business in The Digital Age

The ongoing evolution of Anti-Money Laundering (AML) regulations has resulted in an expansion of the responsibilities of financial institutions. Today, numerous financial entities, including fintech companies, cryptocurrency exchanges, and investment firms, are expected to maintain vigilant monitoring and report suspicious transactions, reflecting the enduring responsibilities that banks have long upheld.

However, despite advancements in technology, mastering AML remains a hurdle for many institutions. False positives, which have historically been a persistent issue for almost every bank, now present a challenge for various other financial institutions as well.

What Are the Different Types of False Positives?

Before delving deeper, let’s explore some prevalent types of false positives encountered by Financial Institutions (FIs) within their AML systems:

• Legitimate Activity Mistaken as Suspicious: Regular transactions with unique patterns or occasional large sums, which might trigger alerts but are actually legal and routine for specific customers.
• Inaccurate Risk Profiles: Misinterpretation of a customer’s risk level due to insufficient or outdated information, leading to false suspicions.
• Geographic or Transactional Anomalies: Transactions to or from high-risk countries or sudden changes in transaction behaviour may trigger false alerts.
• System Glitches or Errors: Technical malfunctions within the monitoring system or data input errors might generate false positives.
• Unusual Account Access: Legitimate customers accessing their accounts from new locations or using different devices might trigger alerts for potential unauthorised access.
• Complex Corporate Structures: Transactions involving intricate corporate structures or multiple subsidiaries can sometimes appear suspicious but are legal and legitimate.

Regardless of its causes, false positives inevitably impose a significant cost on the institution, not solely in terms of the manpower needed for manual reviews, but also in draining resources that could otherwise be allocated to identifying more intricate or severe risks, potentially posing a greater reputational threat to the institution.

Hence, effectively managing these false positives is crucial to streamline investigations and focus on genuine suspicious activities.

How Data Shapes and Influences False Positives

AI and machine learning have indeed revolutionised AML capabilities for FIs. However, according to Ahmed Drissi from SAS Global Fraud & Security Intelligence Practice, the effectiveness of their implementation hinges upon the quality and foundation of the underlying data.

In a recent interview, Ahmed highlighted several common causes contributing to false positives within AML systems:

• Data Quality: Inaccurate, incomplete, or outdated data can trigger false positives.
• Lack of Cross-Referencing: Failing to compare data from various sources can lead to mistaken alerts.
• Missing Third-Party Data: The absence of reliable external data for validation can limit the accurate identification of suspicious activities.
• Algorithm Updates: Regular testing and updating of algorithms are essential to avoid false positives caused by outdated interpretations.

He also highlighted two common challenges encountered by both traditional and emerging institutions. They either rely on outdated solutions which are challenging to replace and prone to generating substantial false positives, or they lack the necessary knowledge, expertise, or experience to implement a comprehensive AML framework.

The Vital Role of Trusted Partners in AML Solutions

Ahmed commented that in order to overcome the mentioned challenges and accommodate their growth, financial institutions often seek enterprise-grade “next-gen” solutions and collaborate with vendors that specialise exclusively in financial crime monitoring.

In this space, he mentioned that SAS holds a unique advantage with over 40 years of experience in applying analytics to business challenges. Ahmed then elaborated on how SAS has the expertise and capabilities to help FIs stay at the forefront of emerging risks and evolving regulations, enabling them to achieve over 90% model accuracy, slashing false positives by up to 80%, and quadrupling the SAR conversion rate.

First and foremost, to ensure the foundation of data quality, SAS uses various techniques to standardise, deduplicate and correct data, and data governance tools that enable institutions to set and enforce overarching rules that control how they collect, manage and archive data.

With a combination of advanced analytics, AI, machine learning and superior scoring algorithms − alongside organisation-specific filtering rules, unique name-matching analytics, multilingual screening capabilities, and consolidated alerts − SAS facilitates entity resolution by scrutinising diverse data sources and references to generate high-quality hits with low false positives.

Best of all, these can work seamlessly with existing AML solutions, so there’s no need for institutions to replace their current platforms.

Enhancing AML Efficiency: Quality Over Quantity in Alert Review

The question is, does adopting a comprehensive AML solution mean your teams will be reviewing fewer alerts? Not necessarily.

In reality, it can potentially identify more anomalous or suspicious activities. However, it also enables a concentrated focus on genuine risks that pose a threat to your institution, so you can direct your investigation efforts more effectively across a significant portion of your portfolio—this is where its real value lies.

Discover how SAS can assist your organisation in enhancing AML efficiency and reducing the occurrence of false positives by clicking here.

(0)(0)