ComplyAdvantage Debuts AI-Powered Fraud Fighting Tool

fraud detection

ComplyAdvantage says it is using artificial intelligence (AI) to ferret out financial crime.

The fraud prevention firm announced Tuesday (May 16) the release of Fraud Detection, a tool that uses machine learning algorithms to detect and prevent scammers amid a rise in payments-based fraud.

“Banks and other financial institutions are constantly playing catch-up,” Oliver Furniss, the company’s chief product officer, said in a news release.

“Fraud Detection gives them an effective new way to monitor the millions of transactions they process every day and stop criminal activity in its tracks.”

According to the release, the tool identifies more than 50 of the most common payment fraud scenarios facing banks and other financial institutions (FIs), including account takeover fraud, authorized push payment fraud (APP) and relationship fraud.

The launch comes days after news that more than $1.5 billion was stolen in the U.K. last year through authorized and unauthorized financial fraud, according to a recent UK Finance report.

As PYMNTS wrote earlier this week, that figure amounts to an 8% drop compared to 2021, but it also indicates that “there’s no debate that the battle to prevent fraud and keep criminals at bay is far from over.”

According to Kate Frankish, who oversees anti-fraud efforts at Pay.UK — the recognized standards body for Great Britain’s national retail payment schemes — the drop in fraud rates is due to efforts by FIs and payments providers. These organizations have spent millions updating their systems to better detect and prevent fraud.

But it remains “a real balancing act,” Frankish told PYMNTS in a recent interview, “because what we don’t want to do as an industry is to slow down and stop valid payments, which only disrupts the payments chain.”

This is where turning to technology such as AI and machine learning can play a crucial role, she said, especially given that nearly 80% of APP fraud — one of the common scams in the country, in which fraudsters impersonate a payee and dupe an individual or business into sending an instant payment — cases begin online, per the UK Finance report.

“Some of the leaps and bounds in AI and machine learning have helped to spot patterns of fraud and money moving around not just in the U.K. but globally,” she noted.

However, AI is “not one silver bullet” which is going to fix everything, Frankish said, pointing to the need for each bank to design its fraud prevention strategy — whether fully manual, tech-based or a mix of both — based on its expertise.