AI-Based Fraud Detection Comes To The Mastercard Global Network
By Christine Kern, contributing writer
Decision Intelligence™ leverages machine learning to boost approvals for genuine transactions.
Mastercard has announced the introduction of Decision Intelligence™, a comprehensive decision and fraud detection service, across its global network. The new AI-based solution uses machine learning technology to increase the accuracy of real-time approvals of genuine transactions and reduce false declines and fraudulent purchases.
The new technology replaces the current decision-scoring products which use a risk assessment rubric based on predefined rules. Decision Intelligence takes a new, broader approach to assessing transactions, creating scores that enable the card issuer to apply the intelligence to future transactions.
According to Business Insider, better fraud detection tools are imperative as card fraud continues to rise, particularly in e-commerce. Between October 2015 and April 2016, e-commerce fraud attacks grew 11 percent, and cost per dollar for digital fraud also increased 12 percent year-over year while physical fraud grew just 3 percent in the same time period.
“We are solving a major consumer pain point of being falsely declined when trying to make a purchase,” said Ajay Bhalla, president of enterprise risk and security, Mastercard. “By using AI technology on our global network, we’re helping financial institutions and merchants improve approval rates — and the consumer experience.”
But better fraud detection doesn’t just mean adding more steps. When efforts add steps or increase manual review, it can lead to cart abandonment and lost sales if customers get impatient. And increased false declines also impact sales. According to LexisNexis, 58 percent of declined transactions were false positives in 2016, and estimated $8.6 billion in lost sales.
Accuracy matters in payment authorizations, according to Mastercard, which says that one in six users experience declines due to suspected fraud costing some $118 billion per year in false decline — thirteen times the cost of actual fraud in the U.S. annually. And false declines also cost retailers customers, since more than one-third say they stop shopping at a retailer after being falsely declined at the point-of-sale.
“We estimate that in the U.S. alone, the value of false declines is more than 13 times the total amount lost to actual card fraud,” said Al Pascual, senior vice president, research director and head of fraud and security at Javelin Strategy & Research. “Applying machine learning to decision-scoring is a new way of creating a positive consumer experience, while also minimizing fraud.”
By reducing false declines, Decision Intelligence helps to lower operating costs for retailers and banks and increase revenue while reducing risk for retailers.