Covering Disruptive Technology Powering Business in The Digital Age

Home > Archives > News > Visa Prevents Approximately US $ 25 Billion in Fraud Using AI
Visa Prevents Approximately US $ 25 Billion in Fraud Using AI


Visa Inc.’s new analysis, Visa Advanced Authorization (VAA) using artificial intelligence (AI) helped financial institutions prevent an estimated $25 billion in annual fraud—making the global payment ecosystem safer for retailers and consumers. VAA is a comprehensive risk management tool that monitors and evaluates transaction authorisations on the Visa global payment network, VisaNet, in real time to help financial institutions promptly identify and respond to emerging fraud patterns and trends.

Highlighting this breakthrough at the Visa Security Summit in Shanghai, China, Joe Cunningham, Visa Asia Pacific Head of Risk said VAA uses Machine Learning capabilities on every VISA transaction.

“This is our interpretation of AI. It has been around for many years. It is one of the most sophisticated implementations of truly powerful machine learning, which is used on every Visa transaction. Every single transaction, as it flows through our network, in a blink of an eye, we look at it and us asses the probability of that transaction is fraudulent. We give it a score, and it goes to the bank with that information. Vast majorities have low scores which are low risk, but some are high risk. The bank will ultimately decide what to do with the transaction. We have calculated that we save the industry every year, approximately $25 billion because of this AI implementation.”

He added that the beautiful thing about this is as more data goes through our network, the more data that flows, the better the predictions. “We learn from the successful patterns, and we learn from the fraudulent patterns. It feeds itself.”

Visa processed more than 127 billion transactions between merchants and financial institutions on VisaNet last year, and employed AI to analyse 100% of the transactions—each in about one millisecond—so financial institutions can approve legitimate purchases while quickly identifying and preventing fraudulent transactions.

Visa Advanced Authorization is a layer of fraud prevention that can help drive down risk and fraud for financial institutions and retailers, and help reduce friction due to false declines for payment account holders. More than 8,000 financial institutions in 129 countries use Visa Advanced Authorization.

Preventing fraud near the speed of light

Visa pioneered using neural networks modelled after the human brain to power its AI platform to identify possible fraud. This delivers faster and deeper insights through previously unknown correlations. Delivered through Visa Advanced Authorization, retailers and financial institutions benefit from:

  • Machine Learning models used for real-time examination of each transaction for indicators of fraud—looking at activities, patterns and more than 500 risk attributes—all in about one millisecond.
  • Risk scoring, which Visa shares with the account holder’s financial institution, where the decision is made to either approve or decline the transaction or flag the transactions for follow up with the account holder.
  • The ability to identify good transactions even when made by new or infrequent shoppers, reducing the likelihood of false declines.
  • Real-time authorisation using integrated, global predictive analytics to identify and prevent fraud.

Visa has kept global fraud rates at historic lows—less than 0.1% —through a multi-layered approach of investing in human intelligence and technology like A.I.; empowering consumers and clients with tools, resources and control to manage risk; and setting governance processes to help businesses and regulators stay nimble.