E-commerce and digital banking allow transactions to be completed in seconds, making it increasingly vital for FIs to use AI to spot fraud quickly and act in real time.

“While AI arms the adversary, it also provides us with symmetric capabilities to defeat them,” Shanon Mclachlan, chief operating officer at tech provider Jack Henry, told FinAi News. “We fight automated threats with automated defense.

“The scale and sophistication of modern AI allow us to achieve vastly superior anomaly and fraud detection across complex banking networks,” he said.

Agentic AI accelerates incident triage and response times, Mclachlan said.

Real-time payments, fraud, losses

There were 266.2 billion real-time payment (RTP) transactions globally in 2023, a 42.2% increase from 2022, tech provider ACI Worldwide reported in its 2024 Prime Time for Real-Time report, the most recent report of complete RTP data. ACI Worldwide predicted that RTPs will increase to 575.5 billion globally in 2028, a 16.7% compound annual growth rate.

RTP growth has led to increased authorized push payment (APP) scams, according to a report released June 30 by business and technology consultancy Capco.

APP fraud occurs when victims are tricked into transferring funds to criminals via real-time or mobile payment systems, according to Capco. It’s hard for victims to recover their money because they did authorize the transaction.

APP fraud losses in the United States totaled about $8.3 billion in 2024, a number that is expected to increase to $14.9 billion by 2028, Deloitte reported in October 2025.

To combat this, FIs need to look at payment risks on their platforms and evaluate their controls, Gregg Henzel, managing principal of U.S. financial crime, risk, regulation and finance at Capco, told FinAi News. From there, AI detection should be overlaid onto the platforms for behavior modeling and trend identification, he said.

Using AI to protect customers

Mastercard this year launched a suite of AI-powered tools to help identify fraudulent merchants. Mastercard Merchant Trust Services, announced in May, is trained to look for early signs of fraud in real time and can handle large transaction volumes more quickly and efficiently than humans, the company said.

“Keeping the scammers from setting up shop will reduce fraud, cut out disputes and provide greater security,” Chief Executive Michael Miebach said on Mastercard’s second-quarter earnings call. “The goal: real transactions for real purchases from real merchants.”

Lloyds Banking Group is rolling out an AI tool in its app to help customers avoid e-commerce scams and shady transactions. Scam Check, announced in June, notifies customers when they might be at risk of sending money to a fraudster via online shopping, a Lloyds spokesperson previously told FinAi News.

Visa is also increasingly using AI for risk and security solutions, CEO Ryan McInerney said on Visa’s fiscal third-quarter earnings call. Visa’s large transaction model, announced June 10, uses AI trained on billions of transactions to improve fraud detection, increase authorization performance and reduce false alerts.

“Early results have shown that it can power up to a five-times increase in fraud value capture,” McInerney said on the call.

Register here for the FinAi Lending Summit, set for Oct. 7-8 in Las Vegas. This inaugural event will include speakers from Fifth Third and Capital One, as well as a fireside chat with Piermont Bank founder and Chief Executive Wendy Cai-Lee.