Money launderers today are much more sophisticated than they were just a few years ago. The methods that criminals deploy for placing funds and layering transactions are complex.
This is in part thanks to the investigative powers of anti-money laundering (AML) programs and technology. Most criminals today understand they cannot simply buy a painting or a yacht or a luxury property for twice its market value to create a new trail of legitimate cash flows.
That’s why anti-money laundering (AML) technology must be sophisticated. It’s our job to remain several steps ahead of the criminals.
Why is AML so important? Is this not simply a regulatory box that banks and financial institutions (FIs) need to check?
Not at all. Money laundering supports all crime, from bank fraud to terrorism to human trafficking. It is what makes these crimes profitable and so attractive to criminal networks.
And it’s big business, too. The United Nations’ estimate is that 2 to 5 percent of the world’s entire GDP gets laundered each and every year. That’s trillions of dollars annually. Money that could otherwise go to public health programmes or famine relief is instead being used to hide delinquents’ tracks.
Stopping money launderers is more than a question of compliance. It’s a moral imperative. And so companies like ours are building better and better technology every day to thwart financial criminals.
Below, we outline how AML tech has revealed new, better strategies for fighting financial crime.
Fight financial crime with anti-money laundering (AML) technology
Anti-money laundering (AML) technology is all of the software, hardware and protocols a financial institution deploys in the fight against financial crime: onboarding systems, sanctions screening, investigations, robotic process automation (RPA) and case management. AML transaction monitoring is the technology deployed to monitor transactions, analyze customer data and flag suspicious activity. When it spots anomalous behavior, it generates an alert that analysts can use in their AML investigations and refer to the relevant competent authorities.
Best-in-class AML transaction monitoring can do the following for financial institutions:
- Give them a holistic and dynamic view of customer behaviors to best understand what usual or unusual activity looks like across the entire customer lifecycle.
- AML systems can prioritize alerts so that investigators can focus on the most worthwhile alerts first.
- Reveal unknown threats and detect emerging risk. Anomalous behaviour detection is how financial institutions remain a step ahead of whatever new methods financial criminals come up with.
- Make the AML system more efficient by reducing false positives.
AML technologies you need to know
IDC predicts that global data creation and replication will experience a compound annual growth rate (CAGR) of 23% over the 2020-2025 forecast period. This explosive growth is being driven by digitization at every level of business and society and the demand for convenience. This has already flowed into financial services and fintech, creating huge databases for AML teams to manage.
It is only possible to fight money laundering today with technology that can scour huge datasets in an instant, then draw meaningful conclusions about the relationships between the transactions and the accounts that it flags.
Here are five of the technologies that make this happen:
AI and machine learning in anti-money laundering (AML)
Artificial intelligence and machine learning hold the potential to analyze those massive datasets efficiently.
AI plays an important role in AML screening and in transaction monitoring. More on both of those below. What’s important to note now is that machine learning supports many of the key processes in an AML system, which makes the system both more efficient and scalable. Additional efficiencies can be achieved by integrating the decisions made from machine learning models into RPA flows, which focuses on automation in the investigative process.
For AML transaction monitoring specifically, machine learning, particularly the newest generation of the technology, allows the AML system to operate on a scoring-based model. This means the AML system can learn from customer data continuously and in real-time as soon as new information is becoming available.
Over time, machine learning helps the AML transaction monitoring system understand more precisely what customers, transaction types and locations are high risk. Then, it builds on what it’s already learned by recognizing new patterns in the data that might appear random to you and I, but actually indicate novel methods of money laundering.
Transaction monitoring tools
Transaction monitoring is one of the key activities in detecting money laundering. These systems must be robust to meet AML compliance, and they must provide banks with a full picture of customer activity so that individual customers or individual transactions can get flagged with precision for further investigation.
Customer identity verification
There are visual identity verification tools that scan a new customer’s ID or other official documents, then help onboard them into the banking system sometimes via video chat. This forms part of your customer onboarding strategy alongside screening technologies.
AML screening software
Screening software enables banks to remain compliant with due diligence requirements and know-your-customer (KYC) laws. AML screening software will check customers against lists provided by national and international agencies. Those lists might flag people who are on watchlists, people who are politically exposed (PEPs) or people who fall under certain international sanctions for FIs to either refuse transactions or mitigate the risk.
This bit of gatekeeping lets the bank understand who their customers really are, and whether their business exposes the bank to risk or criminal liability.
Graph analytics or network analytics tools are often embedded within other AML technologies, and explore the relationships between people. Social media platforms use them to suggest new connections, but banks and financial institutions use similar tools to understand connections between customers.
Graph analytics parse complex, interconnected financial networks to determine whether a single customer has created multiple accounts under different names, for example, or who the third parties might be when a web of money laundering activity is uncovered.
Model governance is as important to AML as artificial intelligence and machine learning are.
Model governance ensures that the complex ML models in an anti-money laundering system are unbiased, transparent and explainable. ML models use new concepts or methods compared to other decisioning systems, so data scientists and engineers have had to devise a framework that makes the model’s operations understandable and explainable by those working in the AML teams.
This way, anyone investigating suspicious activity will have insight into why a transaction or a customer’s account was flagged from an AML transaction monitoring perspective. Additionally, they can meet regulatory requirements around explainability and how models map to regulatory frameworks for AML.
How Featurespace can help your organization fight money laundering
There’s one more important conversation that dovetails off contemporary approaches to anti-money laundering, and that’s a conversation about costs.
When AML becomes a simple matter of regulatory compliance, the risk is in escalating FTE costs. Regulations get more complex and more frequently updated, and AML teams become trapped in a cycle of allocating more headcount to the problem, without solving the greater problem.
Featurespace’s approach to AML reverses that dynamic, and we do this with our Adaptive Behavioral Analytics technology. Adaptive Behavioral Analytics weaves the power of machine learning throughout an organization’s entire financial crime-fighting activities, to improve efficiency and effectiveness in AML transaction monitoring.
It does this by learning the behaviors of each and every customer. It studies when, how and why people make the transactions they do so that the AML system can know immediately whether an outbound $1,500 is a mortgage payment or something suspicious. When AML teams can focus their efforts only on the truly suspicious transactions, they improve outcomes in the fight against financial crime.
ARIC™ Risk Hub
ARIC™ Risk Hub is the tool that brings all of these financial crime-fighting capabilities together.
It monitors customer data for fraud, money laundering and other financial crimes across all transaction types and financial products.
The machine learning innovations our team has developed, including the Adaptive Behavioral Analytics described earlier allow ARIC™ to spot suspicious activity and prioritize alerts, with descriptions as to why a particular event was flagged.
The benefits of ARIC Risk Hub for AML include:
- Reduced false positives for more efficient AML processes
- Investigations focused on worthwhile alerts and higher value activities
- Proactive identification of emerging and previously unknown patterns of activities to reduce risk
- Reduced maintenance workload around tuning and testing of ML models thanks to self-learning capabilities
- Effectively manage risk with a complete and accurate view of exposure
- Overall reduction in the total cost of compliance
Our anti-money laundering technology in action
To see ARIC™ Risk Hub disrupts financial crime, have a look at the video below:
Case study: HSBC
A few years ago, British investment bank HSBC reached out to us for help with transaction monitoring. The organization’s insurance business identified the opportunity to proactively mitigate risk in its transaction monitoring capabilities, and deliver greater innovation to the business.
Our team delivered a cloud-based implementation of ARIC™ Risk Hub, which helped the AML team prioritize alerts. Further, the ARIC™’s self-learning models allowed for the system to be optimized in a more streamlined approach.
This is where the conversation returns to innovations for efficiency in AML. When organizations prioritize innovation they reduce the burden on teams and the need to increase FTEs, and instead position themselves for better compliance down the road, and make it easier for their teams to respond to new money laundering typologies.
Get in touch