A major issue facing financial services organizations is tracking fraud due to money laundering. Trying to track money laundering is an expensive and time-consuming process due to the large volumes of financial data which must be analyzed.

There are financial services compliance requirements for fraud and anti-money laundering (AML) tracking and organizations face steep fines for non-compliance. Typical know your customer (KYC) rule-based processes have a high false-positive rate for identifying money laundering fraud transactions.

Advances in digital banking, online account opening, open banking and cryptocurrency make it even more difficult to track the source of funds. Financial organizations are increasingly using cloud-based, GPU-accelerated artificial intelligence (AI) and machine learning (ML) AML predictive analysis models to identify money-laundering transactions quickly and accurately.

What is anti-money laundering

According to a Thomson Reuters report, “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 ‘cleaned’ of its illegitimate origin and made to appear as legitimate business profits.”

AML compliance tracking requirements

The Bank Secrecy Act requires financial services organizations to analyze customer data looking for fraud including money laundering. The Office of the Comptroller of the Currency (OCC) conducts regular AML examinations. Banks are required to track and report all instances of suspected money laundering to the Financial Crimes Enforcement Network (FinCEN). Fines are issued for non-compliance or mistakes in identifying money-laundering transactions.

Traditional KYC methods are ineffective AML tools

Many organizations use legacy AML Transactions Monitoring Systems (TMS) to identify suspicious transactions that may involve illicit proceeds or legitimate proceeds used for illegal purposes. The predominant TMS technology uses antiquated rules-based systems that rely on structured queries that aren’t precise and evidence suggests that TMS systems generate high false positive money laundering alerts. Investigating false positives is time consuming and can be expensive. According to a Compliance Week article, ”Kroll’s Global Enforcement Review 2022 recorded 55 global money laundering fines issued in 2021 totaling approximately $1.6 billion in value.”

Building an effective AI AML fraud solution

Many financial organizations have legacy infrastructure with CPU-based processing that cannot handle the processing speeds required for AI or ML AML analysis. Moving to a GPU-based infrastructure provides faster processing and training for ML inference models used to locate money laundering transactions.

AI is one of the most promising AML tools available to banking and regulators. A Thomson Reuters report indicates that leading banks are using AI deep learning algorithms such as GANs (generative adversarial networks) and GNNs (graph neural networks) for pattern matching, which is more accurate than rule-based approaches. GANs can generalize from AI training data to identify patterns in transactions that are indicative of money laundering.

According to this Forrester report, 84% of technical decision makers see significant opportunities with Al and believe they must implement Al to maintain a competitive advantage in their industry. However, 51% of the business leaders indicate their organization does not have the right resources to add AI capabilities.

According to a Forrester survey, “What organizations need are prebuilt, configurable AI cloud services. Cloud AI services allow developers to access a depth of AI capabilities via APIs for fueling application innovation without requiring data science experience.” Moving to a cloud-based AI solution that includes prebuilt AI models results in faster deployment time, and gives organizations access to AI models that have been responsibly built and tested.

The “State of AI in Financial Services” survey found that companies are moving to AI analysis for a wide range of transactions. The use of AI for AML and KYC fraud detection was one of the top AI solutions implemented between 2021 and 2022. Using cloud-based, GPU-accelerated AI algorithms can help financial organizations more accurately identify money laundering transactions and prevent fines for non-compliance.

Technology partners provide cloud-based, GPU-accelerated AI AML fraud solutions

Microsoft and NVIDIA have a long history of working together to support financial institutions in providing technology to support AI and ML AML solutions. Using Microsoft Azure cloud and the NVIDIA AI platform provides scalable, accelerated resources needed to run AI/ML algorithms, routines, and libraries.

The partnership between Microsoft and NVIDIA makes NVIDIA’s powerful GPU acceleration available to financial institutions. The Azure Machine Learning service integrates the NVIDIA open-source RAPIDS software library that allows machine learning users to accelerate their pipelines with NVIDIA GPUs. The NVIDIA TensorRT acceleration library was added to ONNX Runtime to speed deep learning inferencing. Azure supports NVIDIA’s T4 Tensor Core Graphics Processing Units (GPUs), which are optimized for the cost-effective deployment of machine learning inferencing or analytical workloads.

Microsoft cloud-based solutions for AML fraud

Moving to the Microsoft Azure cloud solution provides financial institutions with a complete set of computing, networking, and storage resources integrated with workload services capable of handling the requirements of AML algorithm processing. Organizations can use Azure Stream Analytics to do serverless real-time analytics of payments from existing repositories, so that fraud prevention teams can access that data in real-time. Automating processes with technologies like Microsoft Power Platform aids in catching fraudulent activities as they occur.

Summary

Financial institutions are required to track and report potential money laundering transactions and face fines for failure to comply. Historically, financial service organizations used antiquated rule-based Transactions Monitoring Systems (TMS) to address Anti-Money Laundering (AML) and locate money laundering transactions. But TMS systems have high false-positive rates.

Using AL and ML algorithms running on GPU-based cloud solutions can analyze patterns in financial data to accurately identify money laundering transactions. This helps financial organizations save staff time, and helps reduce fines for non-compliance in identifying money laundering transactions.