Key Takeaways from This Blog:
After more than 30 years in banking and roughly two decades sitting in community bank C-suites wearing just about every acronym available, I am firmly in the camp that believes AI should and will become an important part of the banking operating model. However, after banks and credit unions have spent decades telling customers that trust is their competitive advantage, we should avoid automating it away.
American Banker published new reputation research last week that should get the attention of every executive currently building an AI strategy. The article cited The RepTrak Company 's 2026 reputation score for banks which came in at 70.5, still within its “strong” range. However, when the questions turned specifically to Banks and AI usage, only 32% of respondents completely trusted banks to use AI ethically if regulatory guardrails were reduced. Fifty-six percent said they were very concerned about easing AI regulation at banks. Concerns about AI oversight, employment and privacy substantially outweighed the perceived customer-service benefits of the technology.
The good news is the industry has maintained its strong trust reputation, but it needs to be very careful when developing its AI strategies if it wants to preserve it. Community banks and credit unions need to understand that difference before efficiency becomes another excuse to dismantle the very advantage, they keep saying differentiates them. There is a big difference between using AI to make your institution better and using AI to make your institution less human. Trust is developed and maintained through human interaction.
Efficiency Is Not the Same as Fewer People
When financial institutions decide they need to become more efficient, they tend to reach for the same three levers: staffing, technology and marketing. Staffing is expensive. Technology is expensive. Marketing looks discretionary. AI has now added rocket fuel to that thinking. Automate the work to reduce the staff and improve the efficiency ratio.
Unfortunately, randomly reducing employees or inserting technology into a broken operating model does not make the institution nimble. It generally makes an inefficient organization cheaper while also making the customer experience worse.
I have argued before that community institutions need to start efficiency work with the customer, not the expense line. Legacy hierarchies, siloed workflows, redundant technology and disconnected delivery channels create enormous amounts of unnecessary cost. They also waste employee capacity. Before executives eliminate perceived excess staffing and replace it with chat bots, they should ask themselves what human interactions create value along their customers’ journeys, and how they can utilize their staff more fully to reinforce that value.
Fix the Operating Model Before You Automate It
Many institutions still manage their delivery channels as independent verticals. The branch, call center and digital channels each have their own employees, processes, management layers and performance expectations. While the customer sees one institution, we have built three, which creates layers of management cost, inconsistent experiences and underutilized people.
A better operating model allows capacity to move wherever customer demand exists. A branch employee should not become economically irrelevant because lobby traffic slows at 11:30 in the morning. An employee does not have to be centralized to be able to conduct a secure video appointment, handle a digital service request, receive call overflow or follow up. Also, a decentralized branch employee can provide financial education and call a customer whose deposit behavior suggests the relationship may be leaving. The work exists but the operating model often prevents the employee from doing it. Redesign the organization to perceive underutilized branch staff and watch your member journeys improve, while your efficiency ratio plummets.
A well-designed customer journey allows every channel to function as an entry point into the same relationship. Transactions stop being defined by whether they started in a branch, online or through the contact center. People and technology work together around relationship ownership and customer need. Now employees reclaim time for advice, problem solving and relationship expansion; and something else happens as well, trust gets stronger.
Trust Becomes More Valuable as Banking Becomes Less Human
This matters because AI itself is changing. We are quickly moving beyond chatbots that answer questions. AI agents are beginning to operate on behalf of consumers. They can potentially compare financial products, manage subscriptions, execute payments, move money among accounts and make decisions based on parameters established by the customer.
While this is an extraordinary technological development, it is also a trust problem waiting to happen. The American Banker research suggests consumers already recognize the risk. RepTrak found that awareness of a bank failing to use AI ethically could reduce that institution’s reputation score by an estimated 8.5 points. Failure to prevent or adequately respond to AI-enabled fraud carried an almost identical 8.4-point reputational hit. Those are not insignificant numbers. Banks spend years building reputation. AI can jeopardize it much faster.
Sometimes the Human Is the Control
Banking has spent years trying to remove friction and most of that work was overdue. Nobody misses paper forms, unnecessary handoffs or waiting three days for something technology can accomplish in three seconds. However, some types of friction are good. When friction serves as control, it preserves the industry’s primary purpose: the safety and soundness of customers’ assets and information.
I recently wrote about branch employees increasingly becoming fraud investigators, financial educators and protectors of customer assets. That role gives something abstract like “trusted advisor” a tangible form. There is an identifiable person who is accountable to help the customer reach the right outcome. An experienced banker who knows how to ask one more question before the money leaves is adding more value, not less.
Digital competitors can build beautiful interfaces. AI can produce remarkably sophisticated answers. Automated agents will become extraordinarily good at executing transactions. But scale does not automatically create accountability. Community institutions already have that capability sitting inside their organizations. Reorganizing the business model to deploy that capability is the challenge and the opportunity.
The challenge is deploying it better.
Put AI Behind Your People
The strongest AI strategy for a community bank or credit union begins with making each employee dramatically more capable. Use AI to summarize information, surface patterns, eliminate repetitive administrative work and improve decision support. Let technology distribute high value activity where judgment is necessary and handle low-value activity.
This is particularly important for smaller institutions because scale will continue to favor large competitors. Community banks and credit unions cannot win by attempting to replicate the operating model of a $250 billion bank with a fraction of the resources. Community institutions need a nimbler model that uses those assets scale cannot easily buy: proximity, local knowledge, trust and a brand that still feels human. The strategic opportunity is to marry those strengths to a cleaner operating model and better technology; and preserve their hard earned reputation as trusted advisors.