Obsessed with technology?
Subscribe to the latest tech news as well as exciting promotions from us and our partners!
By subscribing, you indicate that you have read & understood the SPH's Privacy Policy and PDPA Statement.
News Categories

AI helps Visa prevent more than US$25 billion in fraud annually

By Raymond Lau - on 16 Jul 2019, 11:15am

AI helps Visa prevent more than US$25 billion in fraud annually

Image source: Visa

New analysis by Visa revealed that the company’s neural network-based AI has helped to prevent approximately US$25 billion in fraud over the past year, including cross-border transactions.

More than 127 billion online and offline transactions were put through the company’s global payment network and analyzed in real time, so that financial institutions can approve legitimate purchases while quickly identifying and preventing fraudulent transactions.

The neural networks are modeled after the human brain and take into account several factors, such as machine learning models and risk scoring. Critically, a balance has to be struck between catching fraudulent transactions and falsely denying legitimate users from paying for their purchases.

Law-abiding consumers should not be punished with bad user experience just to catch a handful of criminals. Yet, security systems should be strong enough to catch these criminals when they do attempt illegal activities.

Thanks to Visa’s AI, this balance appears to be well struck. Since Visa deployed the neural network-based system in 1993, global fraud rate has declined by two-thirds to less than 0.1 percent, even as transaction volume has increased by more than 1,000 percent.

Today, more than 8,000 financial institutions in 129 countries use Visa Advanced Authorization. As more of the world embraces electronic payments, contributing to an expanding pool of personal data attractive to criminals, Visa has committed to further reducing global fraud rates by 50% by 2025.

Source: Visa