AI Accelerates AML Processes in Financial Services


Financial regulators across Europe continue to impose steep fines on banks for failing to comply with know-your-customer (KYC) and anti-money laundering (AML) regulations. At the end of 2021, the Financial Conduct Authority (FCA) fined two of the UK’s biggest banks, HSBC and NatWest, a total of £328.95 million ($436.1 million) for failures in their money laundering processes.

Meanwhile, Members of the European Parliament are calling for cryptocurrencies to be regulated by the European Commission’s Anti-Money Laundering Authority, as illicit organizations continue to find new ways to launder money through the financial system.

Money laundering is a process that criminals use to hide the illegal source of their funds. By passing money through multiple, sometimes complex, transfers and transactions, the money is “cleansed” of its illegitimate origin and made to appear as legitimate business profits.

Technological advancements in areas such as digital banking, online account opening, open banking, and cryptocurrency have made tracking the source of funds and uncovering suspicious patterns and behavior much more data-intensive. resources for financial institutions and their regulators. Traditional methods of automation simply cannot keep up with the increasingly sophisticated ways in which criminal organizations abuse the financial system or the rapidly changing technology.

Artificial intelligence (AI) is therefore one of the most promising AML tools available to bankers and regulators. It can be thought of as the development of computer systems capable of performing tasks autonomously, by ingesting and analyzing huge volumes of data, and then recognizing patterns in that data.

Artificial intelligence tools primarily focus on developing systems capable of performing tasks that would otherwise require human intelligence to accomplish, and at speeds beyond the capabilities of any individual or group. Financial institutions are using AI across their businesses to power applications ranging from risk management for capital markets to virtual assistants for consumer credit customer support.

Fraud prevention is another priority use case for AI in financial services. In reality,
research 2022 carried out by NVIDIA shows that two of the top five use cases of AI for investing are “Fraud detection: transactions and payments” and “Fraud detection: AML and KYC”.

Why is AI such an effective AML tool?

First, AI models and algorithms can consume and synthesize huge volumes of data. These entries are not limited to traditional types of tabular data (i.e., transaction records), but can also include unstructured data (i.e., audio, video, and geospatial). Additionally, AI can ingest the data and act on it in near real-time, allowing authorities to track the movements of bad actors rather than being days or weeks behind.

AI models are designed to detect anomalies in the data patterns they ingest by scoring these behaviors against expected benchmarks, so that bank compliance officers are alerted when potentially harmful interactions may occur. Investigations related to these alerts are often carried out by compliance personnel within banks, and are therefore time-consuming and costly.

Traditional rules-based methods – a common technique before the advent of modern AI – have a high rate of false positives, which means that valuable investigator time is wasted on bad transactions.

Knowledge graphs to transform fraud detection

Large banks are therefore using AI deep learning techniques such as GANs (Generative Adversarial Networks) and GNNs (Graphical Neural Networks). With enough historical financial transaction data, deep learning-based approaches are more effective at pattern matching than rule-based approaches because they can generalize to learn fraud patterns and then use that AI to identify active fraud patterns in the data.

For example, GANs can generalize from training data to identify patterns of transactions that indicate money laundering. That is, after showing some patterns in real-life situations, the corresponding deep neural networks (DNNs) can generalize from the examples to identify similar, modified patterns that might bend the static rules, but are similar enough to the old one. model that they are caught by the DNN. This makes it harder for criminals to avoid detection. They will no longer be able to make small adjustments to the way they launder their money to circumvent a relatively static set of rules.

In addition to GANs, GNNs are another DNN technique that allows investigators to assess relationships between any number of parties to flag potential money laundering behavior. The concept is to construct a heterogeneous graph from tabular data and train a GNN model to detect suspicious transactions and complex laundering activities, as criminals work collaboratively in groups to hide their anomalous characteristics but leave traces of relationships .

Relationships identified by GNN-based models are critical, as AI can identify previously unidentified relationships between entities. With the benefits of capturing relationships, GNNs are more capable of detecting collaborative laundering activities than traditional models.

The positive impact of AI on a bank’s AML operations has been proven by a recent study collaboration between Swedbank, Hopsworks and NVIDIA. In this example, Swedbank and Hopsworks have formed GANs as part of the bank’s fraud and money laundering prevention strategy. Using this solution, Swedbank was able to reduce its false positives by 99% compared to existing rules-based systems and create an estimated increase in investigator efficiency (investigation time) of over 50% in five years.

In addition to leveraging AI for intra-enterprise data ingestion and analysis, federated learning techniques will enable better data sharing across departments, jurisdictions, and enterprises due to its ability to maintain compliance with data sovereignty and privacy regulations.

The greater volume of data available for analysis by AI models will greatly improve the accuracy of the models and make it even harder for bad actors to successfully launder money. In addition, AI technology such as robotic process automation and optical character recognition will make it easier for investigators to analyze documents, creating additional efficiencies and reducing error rates. throughout the process.

Originally published by Thomson Reuters © Thomson Reuters.


About Author

Comments are closed.