Anti-Money Laundering Investigations Require Greater Multi-Sector Cooperation & Improved Decision Intelligence Capabilities
When it comes to anti-money laundering (AML) initiatives, the stakes are extremely high.
Across the globe, money laundering remains a key challenge to address organized crime. This is exacerbated by cryptocurrency, blockchain and the anonymity they afford bad actors who are trying to obscure the source of illicit funds. Rosa van Dam, decision intelligence platform product manager at Cognyte, sheds light on the importance of better multi-sector cooperation and decision intelligence capabilities to improve anti-money laundering investigations.
Money laundering finances organized crime and lies behind illicit activities like tax evasion, drug smuggling, human trafficking and exploitation, terror financing and much more.
The United Nations (UN) has estimated that the amount of money laundered globally per year represents between 2 – 5% of global GDP, or between US$800 billion and US$2 trillion. It’s widely estimated that less than 1% of these illicit proceeds are successfully intercepted and recovered.
When it comes to anti-money laundering (AML) initiatives, the stakes are extremely high. Yet the ability of most financial institutions (banks, credit unions, securities brokers, etc.) is nowhere near what’s needed to combat modern money laundering techniques. The result? The authorities cannot block dirty transactions quickly, uncover criminal networks, seize criminal organizations’ funds, and more.
For most financial institutions, regulatory compliance is the top priority. Accordingly, their technology tools and resources are focused on an increasingly complicated set of global, national, and regional government policies designed to define and protect the global financial system. Financial institutions have traditionally been set up to log and manage records, with a lack of top-down workflows. But this model is increasingly inadequate, as the inaction or neglect of AML investigations leads to severe financial and reputational consequences.
Government law enforcement agencies (LEAs) increasingly rely on financial institutions’ data for their investigations to succeed. Financial institutions depend equally on LEAs due to their reach and ability to navigate legal cooperations across borders. To succeed, organizations in the financial sector and agencies in the public sector (LEAs, tax authorities, AML agencies, etc.) must collaborate more effectively to generate meaningful intelligence and effective risk assessments across organizations for faster, smarter, data-driven decision-making.
But this multi-sector cooperation doesn’t come easily, mostly because the data required to conduct a holistic investigation is based on siloed and disjointed sources. In some cases, this clumsy approach and patchwork integration can create known loopholes for bad actors to exploit.
See More: SOC Audits: Selecting the Best Approach for Your Business
Disparate Data Deluge
The financial regime’s focus on compliance is very costly in terms of the manpower and tools required to meet the regulations. It also seems to be very ineffective in surfacing and tracing laundered money, seizing assets, uncovering criminal networks and bringing them to justice. Financial institutions find it hard to manage even their own data repositories. When you add the external sources required for a successful AML investigation, the data management challenge grows significantly.
Disparate data sources and lack of interoperability make it more difficult to cross-reference databases and to make meaningful sense of the data. For international AML investigations, there’s the added challenge of navigating country-specific privacy and protection regulations, languages and identifiers – with no unified standard.
In addition, data sources are typically fragmented, operating in silos. The structure is not standardized across sources and institutions even when the data is structured. This makes it hard to compare and analyze the data and, eventually, to gather actionable insights.
There’s an additional challenge when synthesizing huge amounts of structured and unstructured data. Until now, the ability to quickly and effectively cross-reference this wealth of data has been largely absent, with little to no meaningful context of the data being analyzed.
When it comes to multi-sector cooperation, a lack of integrated tooling has also hampered organizations’ AML capabilities. There’s also typically no single interface to access or manage disparate data sources across different organizations, with little to no common definition in terms of case management, collaboration and permission tools. This ultimately makes it more difficult to analyze data and qualify leads, impeding organizations from blocking transactions on time, revealing the sources of dirty money, recovering stolen assets and more.
If, in the end, the data isn’t fused and processed correctly, there’s an inherent risk of false positives and incorrect conclusions. Hene, analysts, often can’t piece together key intelligence from the information that’s there.
There are challenges in the commercial marketplace as well. Many AML solution vendors are reluctant to provide transparency into their proprietary machine learning (ML) models, and these models get outdated quickly if not properly maintained. Conventional rule-based solutions based on outdated research are ineffective. Likewise, untransparent ML models prevent users from adding meaningful input to train and improve the model and are thus ineffective as well.
The future of AML Investigation
What kind of software solutions and tools are needed to access and analyze inputs from multiple agencies and financial institutions to perform a comprehensive and holistic AML investigation? Below are some of the key characteristics to look for.
Multi-sector AML cooperation depends on the ability to distill meaningful intelligence across different organizations. Data fusion and ML-driven analytics are crucial for delivering data-driven decisions and risk assessment that leverage disconnected and disparate sources.
Effective data modeling is also essential. It enables the standardization of data structures across disparate sources and allows for querying and finding correlations across all available sources.
The ability to tailor ML analytics to the specific data sources and investigation needs, otherwise known as Pluggable ML, fosters greater platform transparency. It also provides the much-needed flexibility for data scientists to continuously cultivate and share knowledge across a multi-disciplinary team of experts.
The ability to create complex intelligence rules helps generate more meaningful decision intelligence. Therefore, establishing and maintaining flexibility in how rules are built is essential. Operational knowledge should be readily translatable into more effective rules and queries.
Risk scoring and prioritization capabilities are particularly important for AML monitoring. Investigators need more than just the ability to search for suspicious identifiers/indicators. Now, more than ever, they also need an intelligent platform that can help them prioritize AML case investigations based on established risk parameters/thresholds.
Improved multi-sector cooperation between financial institutions and government agencies – leveraging data fusion and ML-driven analytics – will enable analysts to resolve more money laundering cases and reclaim laundered revenue while helping to funnel illicit funds away from organized crime activities.
Have you come across any anti-money laundering investigative strategies that are fool-proof? Share with us on Facebook, Twitter, and LinkedIn.
MORE ON MONEY LAUNDERING
