Banks have spent decades building fraud systems that see one transaction at a time. A charge either looks suspicious or it doesn’t. Fraud rings built their business model around that gap, spreading activity across thousands of payments using stolen cards, mule accounts, shared devices and synthetic identities so no single transaction trips a filter.
The Nilson Report projects that global card fraud losses will reach $403 billion over the next decade, with the U.S. accounting for roughly 42% of those losses despite representing just 26% of total card volume worldwide, according to a press release.
Nvidia’s AI blueprint for financial fraud detection is built around a different idea. Rather than asking whether a single transaction looks suspicious, the system asks whether the people, devices and accounts involved in a transaction are connected to suspicious activity elsewhere. A $47 purchase at a gas station may look completely normal on its own. It looks different if the phone used to approve it also shows up in 60 other disputed charges across three states that week. Or the same card was opened using an address tied to a known mule account.
That is the blind spot fraud rings count on. PYMNTS Intelligence found that unauthorized-party fraud — driven by credential theft and account takeovers — now makes up 71% of all fraud incidents and dollar losses at U.S. financial institutions, up from 48% in 2024. Organized rings move fast precisely because they know the window before detection closes.
Why Transaction-Level Scoring Fails Against Organized Rings
Most bank fraud systems today use a technique called gradient-boosted modeling, a scoring engine that looks at a transaction’s characteristics and decides whether it resembles past fraud. Did the purchase happen in an unusual location? Was the amount out of range for this customer? Did the card get used twice in five minutes in different cities? Those are useful signals for catching individual bad actors.
They are much less useful against a coordinated ring. A ring using 500 stolen card numbers can keep each card’s activity well within normal-looking ranges, making individual transactions appear routine. The Nilson Report found that card-not-present transactions represent the highest-risk category in every world region, precisely because they are easiest to execute at scale with stolen credentials, according to the release.
Nvidia’s blueprint addresses that gap by adding a layer that maps relationships across the data. The technique, graph neural networks, works by building a picture of how transactions, accounts and devices connect to each other, then looking for clusters that share suspicious links. It feeds those relationship signals into the existing scoring model as additional context, so a transaction that scores low on its own can still be flagged if it sits inside a connected cluster of high-risk activity.
PYMNTS reported that Block Chief Risk Officer Brian Boates has pushed banks to move away from reviewing fraud after the fact toward stopping it in the moment. “It’s one thing to find the bad actors after the fact,” Boates said. “But what’s much more effective is investing in more real-time technology.” PYMNTS Intelligence found that 68% of financial institutions have increased fraud detection spending year over year as the problem outpaces older systems.
Real-Time Decisions Inside Live Payment Flows
The challenge with relationship-based analysis is speed. Mapping connections across millions of accounts and transactions takes significant computing power. Doing it fast enough to stop a payment before it clears, typically within a few hundred milliseconds, requires infrastructure most banks have not yet built.
The Nilson Report noted that worldwide card fraud losses totaled $33.41 billion in 2024, and that AI tools have helped the industry build its best fraud-fighting models to date, even as organized crime continues to adapt.
Nvidia’s blueprint uses its Dynamo-Triton inference server to run those relationship checks at payment speed. The system produces a fraud score for each transaction alongside an explanation of which signals drove it, so a fraud investigator can see not just that a transaction was flagged, but that it was flagged because the device matched three others in an active dispute cluster, or because the billing address had been used to open four accounts in the past week. The blueprint runs on Amazon Web Services and Hewlett Packard Enterprise, with Dell Technologies support planned, Nvidia said.
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
Nvidia Wants Banks to Hunt Fraud Rings, Not Just Bad Charges | PYMNTS.com Top World News Today.
Hence then, the article about nvidia wants banks to hunt fraud rings not just bad charges pymnts com was published today ( ) and is available on TOP world News today ( Middle East ) The editorial team at PressBee has edited and verified it, and it may have been modified, fully republished, or quoted. You can read and follow the updates of this news or article from its original source.
Read More Details
Finally We wish PressBee provided you with enough information of ( Nvidia Wants Banks to Hunt Fraud Rings, Not Just Bad Charges .. PYMNTS.com )
Also on site :