CUSO-News---Payments-Report

close

Categories

More Tags

Subscribe to Email Updates

Popular Stories

Pay-by-Bank: Anticipating the Next Wave of Innovations
Understanding Enumeration Attacks and How to Prevent Them
What It Means to Have a World Class NPS
FedNow: Changing the Game for Real-Time Payments
Combatting AI-Powered Fraud wtih AI-Powered Fraud Prevention
Written by Cyndie Martini
on July 11, 2023

In recent months, artificial intelligence (AI) has become very popular, making headlines nearly every day. Part of this popularity was triggered by Nvidia's blowout earnings and upward guidance for continued demand for its AI-based chips.

AI has many applications across the financial services industry. One of them is to detect credit card fraud. Using AI to detect credit card fraud has greatly increased the efficiency of fraud detection. AI fraud detection occurs in real time. But what exactly does it do?

AI requires lots of data so it can find specific patterns. In the AI world, this is called training. Once an AI model is trained, it can be unleashed on the real world to go out and find patterns it was trained on.

When it comes to fraud detection, AI monitors your spending habits, building a pattern along the way. When it detects particular deviations from this pattern, it flags it as potential fraud. The AI-based fraud detection system can also take actions such as:

  • Messaging the cardholder to confirm a purchase
  • Locking the account
  • Notifying a credit card representative who can then contact the cardholder

You can probably imagine how such a system is able to greatly reduce fraud. But it may also trigger false positives. For example, a cardholder tries to make a large purchase, something that has never occurred with this card. In that case, the card might block the transaction.

While the above scenario can be inconvenient, the system will continue learning and improving, reducing the chances of false positives.

Humans could never keep up with the same amount of monitoring. AI can monitor thousands of transactions per second. Not only that, it can make a highly confident decision in real-time to flag specific transactions. If a human had to do the same, the efficiency would drop as monitoring transactions, researching particular transactions, and flagging them would take much longer.

AI fraud monitoring is still in the early stages and will only improve from here. But so will the skills of fraudsters, and the cat-and-mouse game will continue.

 

Let Us Know What You Thought about this Post.

Put your Comment Below.

You may also like: