Next-gen anti-money laundering – robotics, semantic analysis
and AI

For decades, anti-money laundering (AML) has been a hot topic and intensifying regulatory pain point for financial institutions. Requirements for detection and reporting are granular and stringent. Faster payments and the pandemic have increased financial crimes to unheard-of levels. Those who fight money laundering confront spiraling amounts of complexity, risks and suspicious activities.

To make anti-money laundering programs more automated, efficient and effective, financial institutions can use next-gen AML with leading-edge technologies like artificial intelligence (AI) and machine learning (ML).

Such technologies automate manual processes like data consumption and analysis, resulting in faster, earlier event detection and scoring. Techniques like natural language processing manage unstructured data and automatically generate suspicious activity report narratives. Advanced analytics streamlines the decision-making process by triaging alerts and determining which do – or do not – need to be acted on immediately.

Much work has already been done to apply artificial intelligence to low-hanging fruit, such as using robotic process automation to investigate and prepare cases more quickly. Some have augmented traditional rules-based approaches to drive down the rate of false positives and more accurately detect and predict activity worth investigating. Many have also adopted machine learning to supplement – or replace – traditional Boolean logic for detecting potentially suspicious activity.

Today, many firms are taking a hybrid approach with their AML programs – using existing processes in conjunction with next-gen AML capabilities. Over time, this will shift as more firms incorporate next-gen AML capabilities into their routine operations.

Six keys to success with anti-money laundering

• Evolve beyond rules with a hybrid approach. The traditional way of monitoring transactions with rules-based systems can’t keep pace with today’s fast-paced financial crime landscape and evolving AML compliance and counter-financing of terrorism (CFT) regulatory obligations. With AI and ML, you can overcome these limitations. But most firms are not ready to abandon their rules-based systems and fully replace them with analytical models and robotics. Many are taking a hybrid approach instead.
• Take a hard look at your data foundation and legacy systems. Improving data quality is key to building a robust AML program. But data is often fragmented and incomplete and resides in disparate systems across the organization. Mergers and acquisitions create even more siloed systems and databases. To elevate defenses against financial crimes, financial institutions must address internal data inadequacies while integrating internal and external data. This approach gives a holistic view of customers and enhances risk profiling.
• Explore machine learning and artificial intelligence techniques. While it’s tempting to add more people to tackle compliance pressures and increasingly sophisticated threats, that’s not a feasible approach for the long term. Testing the latest technology, such as AI and ML, helps you learn what advantages you can achieve from different methods – such as streamlined processes and faster decision-making. Innovative techniques could help you broaden your coverage of risk while boosting the efficiency of your overall AML program.
• Continuously learn and improve. Today, financial institutions need smarter and more nimble surveillance. Using AI and ML can uncover all types of new and sophisticated financial crime schemes. You should account for new risks in new equations – and that calls for continuous learning. AI can help you automatically deal with these new threats coming through the system – and ML can analyze historical outcomes and automatically adjust thresholds to reduce false positives.
• Establish rigorous model governance. As things change, it’s essential to tune and test your models. Otherwise, they may become less effective over time. Consider the effects of changes in the customer base, or feedback from investigators revealing scenarios that need to be fine-tuned. By continually validating and challenging your models, you can ensure they remain the best for your business process.
• Be prepared to adapt. It’s essential to keep reevaluating your approach so you can adapt as times change. Embarking on a path to modernization can be complicated – requiring time, resources and funding. But today’s technological advancements give financial institutions the tools they need to modernize AML compliance frameworks. Stay prepared – set priorities based on your objectives, and adopt technologies based on their effectiveness and value for your ecosystem.

Innovative financial institutions are reaping the benefits

Here are a few ways that banks are applying these new approaches:

• A tier-one bank deployed a combination of text mining, image recognition and ensemble models that processed 9 million transactions, scanned 25 million documents, automated 200 risk typologies and improved operational efficiency by 25%.
  
• A Tier 2 US bank replaced 10 cash activity scenarios from its transaction monitoring system with a SAS neural network model and tripled SAR conversion rates while cutting monthly work items by 50%.
• A Tier 1 global bank applied a random forest model with 200 trees to nearly 2 billion transactions, and in 10 minutes found 416 suspect entities that, on further triage, resulted in dozens of productive cases.
• Another Tier 1 global bank used machine learning-driven automation to help automate due diligence document review, reducing the effort from two weeks of staff time to less than a minute.
• An Asia Pacific bank turned to gradient boosting and deep neural networks to automate alert review and reduced false positives by 33%.

Next-generation AML has moved to the forefront as the industry goes through massive digital transformation and as regulators keep upping their definition of “reasonable” control and governance. Robotics, semantic analysis and artificial intelligence – particularly machine learning – are central to this evolution.

As technology advances, the barrier for entry has dropped to the point where it is within reach of smaller institutions. You don’t have to have an army of data scientists on staff. SAS is packaging advanced AML data science in a box to automate repetitive manual processes, more accurately detect suspicious activity, and cost-effectively put these capabilities in the hands of more financial services organizations.

Learn more about anti-money laundering solutions from SAS