Overhauling a customer experience is no small task for banks and credit unions of all asset sizes, but new research from American Banker finds that many are turning to artificial intelligence for both nuanced improvements and enterprise-wide operations.
Data and analytics leaders at financial institutions were optimistic about the impact that AI will have on their organizations’ customer experiences over the next few years, with larger players reporting higher levels of progress when compared to their smaller peers.
American Banker surveyed 130 bank and credit union leaders for its 2025 Data-Driven Bank survey. Roughly 60% of respondents said they were either the primary stakeholder or a significant figure in their organization’s data, analytics and/or AI initiatives/plans.
Top findings from the report
All respondents, save community banks, were wholly optimistic about AI helping enhance customer experience.National banks were the most advanced in their adoption of AI tools.Fraud detection and reduced friction were top CX improvements for 2025.AI yielded major results in the financial education of consumers and overall response times.Bankers worry that using AI in interactions with account holders could introduce bias or weaken customer relationships.
Results from the report are highlighted below using interactive charts. Mouse over each section for more detail, and click on the chart labels to show or hide sections.
Artificial intelligence, real impact
Executives at the largest banks were the most optimistic about the effect AI will have on their customer experience journey over the next two years, with all respondents expecting some degree of change.
Those working at banks with between $10 billion and $100 billion of assets were more varied in their expectations: 44% said they anticipate a revolutionary or considerable impact against 66% who forecast a moderate to minor impact.
The only respondents who said AI will have no impact on their customer experience were a small portion (4%) of community bankers.
Experts expect that these disparities are due to shift in the coming years.
Olly Downs, chief technology and AI officer for the data and analytics firm Curinos, said there will be a “rapid acceleration of AI-based capabilities” that reach financial institutions “through proven vendor solutions over the next several years,” including both established providers and new players in the market.
“This will democratize AI capability beyond the largest FIs who can ‘build’ into the tiers of the industry that primarily buy technology,” Downs said.
While it may appear that the larger financial institutions are the core users of AI tools, smaller community-based financial institutions should not be counted out, experts say.
Charles Potts, executive vice president and chief innovation officer for the Independent Community Bankers of America, an industry trade group, said that community banks have been using AI-based tools in some form “since the ’90s.”
“Fraud detection and credit scoring systems were built using what we now call ‘predictive AI,'” Potts said.
“AI tools are not and have not always been only for the largest FIs and community banks will continue to explore reasonable and responsible ways to use these tools to better serve their communities and make their operations more efficient,” he added.
Leading the AI pack
Executives at the largest banks are most optimistic about the impact of AI on CX strategies and the most advanced in adopting the technology.
Roughly 51% of national bankers are moving forward with enterprise-wide adoption of AI tools, followed by 15% of midsize banking respondents and 8% community bankers doing the same, according to the survey. No credit union leaders surveyed said they were aggressively pursuing AI adoption.
Among midsize and regional bankers, a third (32%) are exploring small-scale uses of AI for specific tasks, another 29% are engaged in small AI pilots. Most community bankers and credit union leaders are either in the exploratory phase or deploying small pilots.
President Trump’s “desire for the U.S. to be number one in the AI race” has experts like Dylan Lerner, senior digital banking analyst for Javelin Strategy & Research, suggesting that “lax regulations, federal grants and tax breaks or partnerships between public and private institutions” are helping to spur AI advancement among banks and credit unions.
“The pace of AI adoption in financial services will accelerate over the next three years simply because it is becoming more accessible,” Lerner said. “However, it will likely result in the commoditization of basic AI models for underwriting, chatbots, back-end virtual assistants, compliance and so on.”
“The operational efficiencies and improved customer experiences from AI that are competitive advantages today will swiftly become table stakes tomorrow,” he added.
But just as adopting these tools is a key step in banks’ AI plans, so too is measuring their effectiveness.
Alex Sion, head of the financial services vertical at AI solutions provider Blend360, said that many institutions misstep in their AI plans by lacking “a clear ‘before AI’ and ‘after AI’ transformation vision and related set of metrics to measure progress.”
First on the list…
Bankers and credit union leaders are prioritizing improvements to fraud detection, ease of platform use for consumers and security and identity verification in the coming months.
Roughly 68% of those surveyed held that fraud improvements were highly important, followed by 60% who said the same for increased ease of use for portals and 57% for enhanced security and ID verification.
Other key CX improvements include more accurate responses for consumer queries, integrated customer experiences across all channels, faster response times and reducing friction in the onboarding process.
For strategic advisors like Wyatt Mayham, chief executive and cofounder of Northwest AI Consulting, the ongoing wave of “AI-as-a-Service (AIaaS) companies” coming to market are advancing what he says is the “cloudification” of AI, making improvements like these easier for institutions to do with outside fintech partners.
“A $500 million-asset credit union can’t outspend JPMorganChase on R&D, but it can use these off-the-shelf AI tools to automate back-office tasks and free up its staff to do what they do best: build actual human relationships with their members,” Mayham said.
Where is AI heading next in the customer experience journey?
Some uses for AI in customer experience go beyond products, as evidenced by the 77% of respondents that saw improvements in how financially literate customers were following the addition of AI.
Other business aspects that have seen gains after adding AI support include response times (87%), increased ease of platform use (83%) and enhanced security and identity verification (85%).
Agentic AI, wherein a large language model is granted a degree of autonomy to perform tasks on its own, is an emerging technology among banks and credit unions that promises to bring a greater degree of change when compared to traditional forms of AI.
Use cases range from interpreting unstructured data sets to modernizing legacy coding, streamlining call center operations and more, said Phil Andriyevsky, AI and data leader for Ernst & Young’s financial services division.
“Financial institutions are piloting agentic AI in high-impact areas like compliance monitoring, sales prospecting, credit decisioning and internal research. These applications are designed to operate within clearly defined boundaries, surfacing insights, drafting alerts and generating outreach while ensuring that final decisions remain in human hands,” Andriyevsky said.
Bankers’ fears of AI running amok
When it comes to top AI concerns for banks revising their customer experiences, regulatory worries aren’t as dominant as one might expect.
Unauthentic messaging, diminished relationships with customers and introducing bias or unfairness into customer interactions were all significant concerns identified by banking and credit union leaders.
Misleading information and uncertainties around customer privacy and security were other hangups identified by those surveyed.
While larger financial institutions might have the scale and resources to address these concerns, smaller community-based banks and credit unions are oftentimes not as technologically adept, according to Zor Gorelov, senior advisor for advisory firm Klaros Group.
“Community banks and credit unions often operate with lean IT teams and limited in-house AI expertise, [and] as a result, they’re increasingly dependent on third-party vendors. Without robust evaluation frameworks, governance structures and decision-making processes in place, these institutions risk wasting time, money and eroding trust in the potential of AI to transform their business,” Gorelov said.
“Establishing clear oversight is not optional, it’s essential for making sound, strategic technology investments,” he said.