For many credit union leaders, the questions surrounding artificial intelligence have changed. Not that long ago, the conversation centered on basic education: What is AI, and where do we begin? Today, more leaders are asking a harder and more consequential question: Where can AI create meaningful impact for our credit union?
That shift matters because access to AI is no longer the greatest hurdle. Many institutions already have AI embedded within vendor products or have made generative AI tools available to employees. The harder work is determining where those capabilities belong, how they can support existing workflows, and who will remain accountable for the results.
That is an encouraging sign of progress, but it is also where the work becomes more complex.
Moving beyond exploration requires more than purchasing a tool or identifying an appealing use case. Credit unions need a structured way to evaluate readiness, uncover opportunities, set priorities, and prepare employees for change. Otherwise, individual experiments can remain disconnected from organizational strategy, and early investments may produce activity without producing lasting value.
Put AI on the Leadership Agenda
AI should not live exclusively within IT or depend on a handful of enthusiastic employees as it can affect operations, workforce responsibilities, member experiences, data practices, and risk exposure, it deserves attention at the executive and board levels.
Clear ownership of how AI is used is an important first step. “AI needs an owner at the leadership table—someone responsible for guiding education, identifying and prioritizing opportunities, and coordinating governance across the organization,” says Bob Marsh, Chief Growth Officer of OnTrac AI.
The right person may come from operations, strategy, organizational transformation, technology, or another cross-functional role. More important than a particular title is the ability to connect business strategy with how work is actually performed. This leader should be curious about the technology, credible with executives and employees, comfortable guiding organizational change, and able to bring the right voices into the conversation—including IT, information security, compliance, risk, human resources, and the employees closest to each process.
Your AI leader does not need to make every decision or possess expertise in every discipline. An executive sponsor can provide authority, while a cross-functional working group or AI council contributes the operational, technical, and risk perspectives needed to evaluate opportunities responsibly. The leader’s role is to keep those efforts connected and moving toward a shared objective.
National research reinforces the importance of visible leadership. Gallup found that employees who strongly agree their manager supports AI use are twice as likely to use AI frequently. Access to technology may open the door, but leadership support helps employees understand how—and why—to walk through it.
Leaders also need a shared definition of what the institution is trying to accomplish. Is the priority to create capacity for an overextended team? Improve a fragmented workflow? Help employees find and use information more efficiently? Support growth without proportionally increasing administrative work?
When the desired outcome is clear, the credit union can evaluate AI as one possible means to reach it—not as the strategy itself.
Start With the Problem, Not the Technology
Before designing an AI-enabled future, a credit union must understand how work gets done today. That begins with employees.
The people performing a process know where delays, repetitive steps, workarounds, and information gaps occur. Involving them helps leaders identify real opportunities rather than making assumptions from a distance. It also builds trust by positioning employees as contributors to the transformation—not recipients of a decision already made.
Current-state process mapping can then reveal an essential distinction: which steps could be supported by technology, and which require human judgment, decision-making, or review?
Some opportunities may call for AI. Others may be better suited to traditional automation, process improvement, or additional training. A readiness assessment should also consider data quality, system integration, security, governance, leadership alignment, and the organization’s capacity to maintain what it implements.
These foundations do not need to be perfect before a credit union begins. In fact, controlled experimentation can expose gaps that policies and planning alone will not reveal. But the institution should understand its starting point and create a safe environment in which teams can learn.
Turn a Long List of Ideas Into Priorities
Once employees and leaders begin imagining possibilities, ideas can surface and multiply quickly. The challenge is deciding which ones deserve attention first.
“A useful evaluation looks beyond technical feasibility. Each opportunity should be considered according to organizational impact, implementation effort, risk, readiness, and likelihood of employee adoption,” explains Marsh. “For instance, a project with modest technical complexity and clear employee support may create more near-term value than an ambitious initiative that depends on unavailable data, immature processes, or extensive change.”
Even a test that reveals a poor fit can be valuable, helping the credit union sharpen its evaluation criteria and direct resources toward stronger opportunities.
For many credit unions, an internal, employee-facing process may also provide a more manageable starting point than a complex member-facing application. A contained initiative allows the organization to test accuracy, establish oversight, gather employee feedback, and learn what responsible adoption entails before expanding into higher-risk areas.
This is also why starting small should not be confused with thinking small. A focused first initiative can help the organization establish governance, build employee confidence, and learn how to measure results. Those lessons create a stronger foundation for more advanced work later.
The objective is a roadmap—not a collection of disconnected pilots. Leaders should be able to explain why an opportunity was selected, what success will look like, who will remain accountable, and how the work supports the credit union’s larger strategy.
Prepare People, Not Just Technology
AI adoption is ultimately an organizational change effort—and employees need practical education, clear expectations, and a voice in the process.
“The more employees are involved in identifying challenges and developing solutions, the more likely they are to embrace the change,” Marsh explains. “They become part of authoring the solution rather than simply being asked to adopt it.”
That involvement should begin early, particularly because some employees may already be experimenting with AI independently. Clear guidance, approved tools, and practical training can channel that interest into safer, more productive use.
Those closest to the work have an important role to play, Marsh notes. They can help map processes, uncover friction points, design improved workflows, and establish appropriate checkpoints for human review. “Along the way, employees need to understand both what AI makes possible and when its outputs must be questioned or verified.”
Efficiency and oversight do not have to compete. Used thoughtfully, AI can support appropriate tasks while giving people better information to make the judgments and decisions that should remain with them.
From Interest to Measurable Action
Through its partnership with OnTrac AI, MDT is extending its Transformational Consulting capabilities with deeper AI readiness, strategy, and implementation expertise. The shared goal is to help credit unions align their people, processes, and technology; identify the AI opportunities that matter most; and build a practical path from exploration to measurable value.
That path will look different for every institution. Some credit unions need greater AI fluency first. Others are ready to assess workflows, prioritize opportunities, or build an implementation roadmap. The right starting point depends on the organization—not on how quickly another institution is moving.
Before investing in disconnected tools or isolated use cases, credit union leaders should take the time to understand where AI can create value and what their organizations are prepared to support.
“The goal is not to adopt AI everywhere at once,” Marsh concludes. “It is to understand where the technology can create value and build the organizational readiness to support it.”
That clarity is not a delay in progress. It is what makes meaningful progress possible.