Fighting Money Laundering More Effectively with AI-Enhanced Tools

Artificial intelligence is a set of statistical techniques that enable computers to see relationships, make inferences, and predict scenarios based on patterns drawn from large amounts of data. Financial services companies use AI techniques to automate back-office processes, including credit card fraud prevention, personalizing product offers, formulating recommendations for sales teams, and combating money laundering.

Traditional rule-based systems, which look for warning signs of criminal activity or suspicious transactions based on pre-programmed patterns, are giving way to AI-based systems that can detect the behavioral characteristics of money laundering. Previously, anti-money laundering software looked for red flags that might indicate criminal activity, as well as supplementary information such as a banking customer appearing on an international sanctions list, bank deposits just below the threshold requiring reports to be sent to the government, or transfers of amounts from an account similar to those recently paid.
The difficulty is that criminals use evolving tactics to launder their proceeds through what appear to be legitimate financial transactions. In addition to setting up shell companies to make ownership harder to trace, they invest in existing businesses that do most of their business in cash and then inflate their revenues. They also deposit their money in small amounts across multiple financial institutions and route money to countries with lax regulations. This means traditional anti-money laundering methods are often ineffective, while generating a very high number of false positives that can cost banks tens of millions of dollars per year.

Artificial intelligence can help banks reduce their regulatory compliance costs by detecting more precise changes in customer behavior and adapting to new risks as they arise.
Software helps combat money laundering by finding previously hidden risks and reducing the number of false positives that AML teams need to investigate.
Banks still maintain a similar number of reports to authorities regarding legitimately suspicious activities.
The cost of inefficient processes is high, as global regulatory fines for failing to stop money laundering are increasing.
Money laundering refers to how individual actors or criminal groups inject the proceeds from their illegal activities into the global financial system to give the impression that these proceeds were earned legitimately. US banks spend about $25 billion per year to combat money laundering, and fines imposed on banks worldwide that fail to prevent it exceeded $6 billion in 2023.
Criminals are finding increasingly sophisticated ways to evade controls, while banks struggle to spot actual money laundering actions: indeed, the vast majority of alerts their tracking software generates are actually related to benign transactions. These false positives consume money and effort.
Today, financial institutions are beginning to supplement or replace predefined rule-based anti-money laundering (AML) software with more sophisticated AI-optimized software. This software is better at detecting hidden patterns in transactions and relationships between people and companies, it examines suspicious activities more carefully, and it more effectively classifies clients based on their money laundering risk level. This can result in fewer false positives, better protection against illegal actors and regulatory fines, and reduced compliance costs.
AI-based systems can detect hidden transaction patterns among networks of people, compare behaviors with those historically common for a company or its peers, assign risk scores to clients based on their past activity and Know Your Customer (KYC) information, and sort events to close low-risk investigations. Fraud detection for transactions, electronic payments to suppliers, AML, and KYC are among the top five use cases for AI in financial services, according to