Money laundering is big criminal business worldwide. It’s estimated that the global money laundering business is worth somewhere in the region of $2000 bn, of which only around 0.2% is detected. In response, many financial institutions (FI) are starting to use AI to detect small anomalies within a large amount of data.
Anti-Money Laundering (AML) is a particularly challenging area of regulation for banks – even more so for large, geographically diverse institutions. Failure to have adequate processes can result in massive regulatory fines. 2019 was an exceptional year for money laundering fines; regulators hit banks with a near record $10 billion worth of fines. There has been number of reported cases related to Anti money Laundering in recent past, likeCapital One Bank that was fined $100 million over Money-Laundering controls andUBS that was hit with $15 million in fines for AML problems.
At the same time, cases of fraud have grown exponentially in recent years due to an increase in e-commerce and online transactions. A staggering $24.26 billion was lost to payment card fraud worldwide in the past 12 months. According to the same source, global card losses are supposed to grow by almost $10 billion over the next three years.
FIs Turn to AI
No wonder financial institutions are investing in AI for fraud prevention. Regulators are also encouraging FIs and banks to use the power of AI and machine learning to detect suspicious activity. AI’s primary benefits for the banks is to help increase the efficiency of financial crime investigations and improve the financial institution’s risk management area.
The vast majority of the banks are using fraud detection systems to analyse clients’ behaviour, location, and buying habits and trigger a security mechanism when something seems out of order and contradicts the established spending pattern.
Additionally, banks employ artificial intelligence to reveal and prevent money laundering. As the act of laundering money is harder to detect at single transaction, single bank level and a wider network of activity, AI and machine learning algorithms is a key to detect this type of criminal activity wherein silos across entities are broken down. These solutions help to cut the costs of investigating the alleged money-laundering schemes by recognizing suspicious activity.
Reduction of Fraud-Related Losses
Cognizant has helped several FIs and banks to combat fraud with AI. For example, a multinational FI asked us to develop a solution that would improve fraud prevention and reduce the reputational impact of false positives. Each month, the FU experienced fraudulent credit card transactions with a value of $75 million that escaped detection and were processed (while the company successfully recovered about $45 million of this amount).
We built a new solution based on open-source technologies and using the latest machine learning (ML) algorithms. Transactions are now run in real time against multiple offline neural network models, which are deployed based on performance metrics. This automated solution was able to handle 4,000 transactions per second and provide a fraud score 99.999 times out of 100.
The solution reduced fraud-related losses by $60 million per year through capturing new fraud patterns, improved customer satisfaction by lowering the rate of legitimate transactions that were declined and reduced operating costs. The solution also provides the ability to rapidly deploy new fraud-detection models. The second phase of the project is now underway and is expected to capture substantial additional savings.
Please visit this page to learn more about Cognizant's Banking & Finance offerings.