Fraudsters often obtain millions of payment card numbers through spyware, malware and other stealthy actions, such as card skimming. They then post some of the cards' 16 digits on illegal websites to sell this data to other criminals. However, payment company MasterCard has announced that it is exploring new ways to use generative artificial intelligence to detect potentially compromised cards more quickly.
The company now has improved capabilities to predict the full information of compromised cards on its network, allowing banks to block them faster than before.
MasterCard's new generative AI-based prediction technology can protect future transactions from the latest threats.
This approach doubles the detection rate of compromised cards, reduces false positives in the detection of fraudulent transactions with potentially compromised cards by up to 200% and increases the speed of identifying merchants that are at risk or have been compromised by fraudsters by 300%.
"Until now, fraudsters might have thought they were operating in the dark, trying to launder the card details of millions of unsuspecting victims. Thank you to our world-leading cyber technology, we can put the puzzle back together, boosting the confidence of banks, their customers and the entire digital ecosystem," said Johan Gerber of MasterCard in a press release.
The payment company can alert banks faster and more accurately if a card has been compromised. This way, the card can be blocked and reissued. MasterCard can also monitor transaction attempts with the compromised product to prevent fraud.
"This latest enhancement, which uses innovative generative AI techniques, strengthens MasterCard's security solutions. Cyber Secure, available since 2020, uses integrated technology to create transparent cybersecurity intelligence based on the online profiles of banks and merchants in the payment ecosystem, including details of suspected compromised cards," the statement reads.
Millions of dollars invested
In February, El Economista reported that MasterCard has invested over 7 billion dollars in combating fraudulent payments and fraud over the past five years.
The company combines tens of thousands of variables with its proprietary advanced artificial intelligence model, Decision Intelligence Pro.
Ajay Bhalla, president of MasterCard's Cybersecurity and Intelligence business unit, told CNBC that the generative AI was developed from the ground up by the multinational company's cybersecurity teams.
He also acknowledged that due to the nature of the business, MasterCard has all the transaction data necessary to detect patterns of behavior and prevent potential irregularities.
The algorithm was trained using data from 125 billion transactions processed annually across the company's network.
In just 50 milliseconds, the company can use its algorithms to set up a scoring system to investigate potential payments that deviate from behavioral patterns.