Banks Deploy AI to Trace Stolen Funds After Instant Payments Clear

Real-time payments have a nasty habit of making fraud final. Once an instant transfer settles, there is no recall button, which is why banks are now pushing AI from “should we approve this?” into “where did the money go?”

1. PYMNTS Intelligence found that 40% of financial institutions lost more money to fraud last year, while 38% experienced higher fraud volumes. Scams now account for 23% of fraudulent transactions reported by financial institutions, after a 56% year-over-year rise, and the share of dollars lost to scams increased by 121%.
2. Authorized push payment (APP) fraud is the cleanest example of why payments teams hate speed when they are on the wrong side of it: the victim authorizes the transfer, the credentials are valid, and the transaction clears before anyone realizes the money is being walked through a chain of accounts. In the U.K., APP fraud losses rose 19% to 576.4 million pounds (about $774 million) across cases last year, with 66% of cases beginning on online platforms, according to a June 15 report by PYMNTS.
3. Nasdaq Verafin has expanded its Agentic AI Workforce with two role-based agents: an Agentic Fraud Analyst and an Agentic AML Analyst. The Fraud Analyst will initially triage unusual ACH activity, while the AML Analyst will focus on cash structuring alerts before expanding to flow-of-funds analysis and unusual international transactions. Both are set for general availability in the third quarter of 2026.
4. The useful bit for banks and PSPs is not the marketing label, it is the workflow. Fraud AI used to stop at authorization decisions; these agents are aimed at the post-clearance phase, where the question becomes what the transaction connects to and where the funds went. Nasdaq Verafin said its platform is already used by more than 650 financial institutions and runs on a consortium data network spanning more than 2,800 institutions, giving it cross-institution counterparty risk visibility rather than a single-bank view.
5. Nasdaq Verafin said its existing agents have already cut alert-review workloads: the Agentic Sanctions Analyst reduced alert review by up to 90%, and the Agentic EDD Analyst cut enhanced due diligence review time by up to 50%. For institutions dealing with instant rails, that matters because the investigative bottleneck has become the product.

In practice, the payment rail is no longer just a transport layer. It is also the starting point for an investigation stack, and that is where AI vendors are now competing: not on stopping the first transfer, but on reconstructing the path after the money has already moved.