Google LLC’s cloud unit today debuted a new service that uses artificial intelligence to help financial institutions detect money laundering.

The service is known as Anti Money Laundering AI, or AML AI for short. 

Banks historically used so-called rules-based detection systems to spot illicit financial activity. In such systems, users manually define what type of transactions should be flagged. Automation algorithms then compare transactions against the user-defined detection criteria and generate an alert if there’s a match. 

When a new money laundering tactic emerges, traditional detection systems must be manually adapted to spot it. The need to perform configuration changes by hand makes the process slow in some cases. That can make it difficult to address new money laundering tactics in a timely manner.

According to Google Cloud, its new AML AI service offers a more efficient alternative. It removes the need for a bank’s employees to manually define detection rules and instead relies on machine learning to spot money laundering. However, banks can reportedly customize some aspects of the workflow if needed. 

Making the money laundering detection process more automated can increase efficiency. The search giant says that HSBC Holding plc, an early adopter of AML AI, started identifying two to four times more suspicious activity after deploying the service.

Because of their accuracy limitations, traditional rules-based detection systems can not only overlook suspicious activity but also flag legitimate transactions as fraudulent. Such incidents are known as false positives. Google Cloud says AML AI helped HSBC reduce false positives by more than 60%. 

Reducing alert volumes saves time for a bank’s employees. Instead of investigating false positives, workers can focus on reviewing suspicious activity. According to Google Cloud, reducing false positives can also improve the customer experience because flagging a legitimate transaction as suspicious requires banks to unnecessarily hold it up for review.

When AML AI spots suspicious activity, it generates an alert and highlights items of interest. The service lists the transactions that led its AI algorithms to raise the alarm. AML AI also surfaces other types of contextual information to help speed up investigations.

HSBC is one several financial institutions that have adopted the service. Other early adopters include Banco Bradesco S.A., one of the largest banks in Brazil, and Denmark-based online bank Lunar. 

“Building on our commitment to bring AI-powered innovation to the financial services industry, we are launching Google Cloud’s AML AI to help financial institutions more accurately and efficiently identify AML risk while enhancing business operations and governance,” said Google Cloud Chief Executive Officer Thomas Kurian.

AML AI is the latest in a series of industry-specific offerings that Google Cloud has introduced for the financial sector over the past two years. 

Previously, the company rolled out a reference architecture that makes it easier for banks to link their infrastructure to the SWIFT international payments network. Earlier, it debuted a cloud service called Datashare. The latter offering makes it easier for financial institutions to access the market data they use to inform investment decisions. 

Image: Google