Most FIs use AI in some capacity, but many continue to place their trust in tried-and-true legacy security systems.

“If your legacy tools don’t involve the newer tools that [fraudsters] are using, then I think the obvious answer is that you’re exposed to those things,” Jose Caldera, chief executive of tech firm Yanez Compliance, told FinAi News.

FIs must be knowledgeable about the newest technologies and use them to remain secure, he said.

This is especially important due to the broad risk posed by a possible cyberattack on an FI, Shanon Mclachlan, chief operating officer at tech provider Jack Henry, told FinAi News.

“FIs represent systemic risk,” Mclachlan said. “An outage or data breach can quickly cascade across broader markets, which dictates exponentially higher operational resilience expectations.”

Systems must integrate “modern, frontier AI tools directly with legacy core banking systems, ensuring comprehensive coverage without disrupting daily operations,” he said.

Foundational work

FIs are increasingly implementing AI, Tracy Moore, director of thought leadership and regulatory affairs at fintech Fenergo, told FinAi News.

The Financial Services State of the Nation 2026 report, released Feb. 10 by Finastra, found that 90% of FIs across several countries are using AI in some capacity. Of the remaining 10%, 8% are exploring AI use and only 2% are not using and have no plans to use AI.

But AI implementation requires significant foundational work, which might be frustrating for FIs that overhauled their systems to implement cloud just a few years ago, Moore said.

“Do they have the data in place?” she said. “Do they have the infrastructure in place to add advanced technology?”

When implementing AI, FIs must avoid siloed systems and ensure consistent practices across business segments, Moore said.

When implementing AI for security, FIs have two options: They can replace their legacy systems or they can layer AI capabilities atop their legacy systems.

“The fortunate thing is, because technology is advanced, it’s not the same to upgrade a system [as it was] in the old days with the [on-premises] and configuring and testing and rolling out,” Moore said. “It’s easier for banks to keep up with technology if they have the infrastructure in place.”

Need for speed?

AI is enhancing the scale and sophistication of cyberattacks and fraud.

“Consequently, cybersecurity has become an urgent business priority with high visibility at the executive and board levels, driving faster investment decisions,” Christiane Ohlgart, chief financial officer at cybersecurity and networking firm Fortinet, said on the company’s July 29 second-quarter earnings call.

“As organizations move from AI experimentation and early adoption toward broader deployment, they require security platforms capable of protecting AI models and datasets while securing large volumes of east-west traffic and enforcing zero-trust segmentation across distributed AI workloads,” she said.

Texas-based Frost Bank has been taking a cautious approach to AI implementation. The $53.9 billion bank hired London-based software developer Mimica this year to explore AI efficiencies for the bank, according to a Jan. 21 LinkedIn post from Mimica.

“AI is helpful in back-office applications to make our processes more efficient,” Bill Day, senior vice president of corporate communication at Frost, told FinAi News. “That includes measures to help keep our customers’ funds and data secure.

“We would describe our implementation as intentional,” Day said.

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.