This article was originally published in CUInsight on November 16, 2026.
By Jay Kiew, Innovation and Change Navigation Strategist
It's a Tuesday morning in November. Maria has worked the member service line at her credit union for nine years. In that time she's tackled two core conversions, a new digital banking platform and a new phone system. Today's all-staff email says AI is coming next.
She doesn't hate change. She's tired of it. And her members can hear it in her voice.
Multiply Maria by every branch and back office in the movement and you get the real challenge behind AI in credit unions. According to the January 2026 Credit Union Tracker from PYMNTS Intelligence and Velera, about 42% of credit unions have put AI to work in specific areas, but only 8% use it across multiple parts of the organization. More than 8 in 10 call integration with existing systems a major obstacle.
Integration is a real challenge.
And so is the human side, which can sometimes weigh on us heavily.
Resilience helps a team survive one more rollout. Flourishing means people come out the other side with more energy, more connection to each other and more time for members than they had going in. Do you feel like your teams are resilient ... or flourishing?
Four change mindsets
Every credit union has all four in the building, sometimes in the same meeting. I call it the Change Mindset Matrix, and it maps our mindset with two simple questions: do you see change as a threat or an opportunity, and do you resist it or adapt to it?
The Defender (threat-focused, resistant): "We've served members this way for 70 years. If it ain't broke, don't fix it." Defenders often care deeply about members and colleagues. They assume protecting the current system is the same as protecting people.
The Observer (opportunity-focused, resistant): Sees the potential but waits for another credit union down the road to prove it first. Often they've been burned by a rollout that went sideways, and caution is how they protect their team.
The Saboteur/Salvager (threat-focused, adaptive): Says yes in the meeting, but when the pilot gets no budget and no dedicated staff time, they actively work against change.
The Fluent (opportunity-focused, adaptive): Assumes disruption carries opportunity and asks "Why not us?" instead of "Why us?" Fluent teams are curious, and curiosity is where joy shows up at work. Credit unions have a head start here. Cooperation among cooperatives is one of our founding principles, so fluent leaders share what they learn instead of guarding it.
Notice that three of the four are reactions to how change has been done to people.
But nobody gets moved to fluent through an all-staff meeting.
Leaders get them there by changing how change happens.
Which one were you the last time someone pitched an AI pilot?
Which one is your frontline team right now?
Three habits that build fluency (and make room for joy)
1. Discover what's possible (strategic foresight)
Ask your frontline staff one question: what part of your week drains you most? The answers usually point straight at AI's strongest use cases, which share three traits: high volume, repetitive and rule-based. Routine member questions like card locks, payoff amounts and address changes. Loan document intake. Call notes for the next person who picks up the member's file. First-pass triage on BSA/AML alerts. AI handles the first pass. Your people review it and own the decision.
2. Design the future (design thinking)
Redesign the workflow with the people who do it, not for them. Map an auto loan from application to funding, from the member's side of the counter, and mark which steps need human judgment and which just move data between systems. The people closest to the work will spot that line fastest.
Verity Credit Union in Seattle did a version of this with AI-automated underwriting through Zest AI. In a 2024 release, Verity reported automated approvals up 100% for auto loans and 84% for credit cards, along with higher approval rates for applicants 62 and older, women, and Black, Hispanic and Asian Pacific Islander applicants. Faster answers, more members hearing yes, and lending staff freed up for the conversations that need a human.
3. Differentiate from the standard (applied innovation)
A chatbot on the website is fast becoming table stakes. Differentiation comes from what you do with the time AI gives back. If Maria gets two hours a day back from repetitive calls, does that time go to more of the same, or to the member who needs a conversation about consolidating debt before it turns into a charge-off?
That second conversation is why most people chose to work at a credit union in the first place. Give it back to them and you're building the kind of culture people want to stay in. The same goes for the back office, where finance teams can now work with Microsoft Copilot and Claude inside Excel to close the month, refresh board reports and run variance analysis, with calculations you can trace back to the source.
Where to start
Pick one low-risk workflow your team dislikes. Document intake. Call summaries. Monthly variance commentary. Don't start with anything a member or examiner will see unreviewed.
Before you begin, define the human review step, keep member data inside approved and secured systems, and measure time saved and quality gained. Then add two questions only a people-first cooperative would ask: did the member's experience get better, and does the team have more energy than before?
Celebrate the first win out loud. Small, visible wins rebuild trust faster than any change management plan. Then tell the credit union down the road. That's community, and it's a cooperative advantage no fintech can copy.
Our goal should be to redesign work responsibly, so the time we get back flows to the people who make a credit union a credit union: our members, and the team that serves them.
I'll go deeper on this at CUInsight Mini-Con: Winter 2026 on Wednesday, November 18, opening the virtual event at 1:00 p.m. ET with my keynote, "Flourish: Inspiring Resilient Leadership and Joyful Culture through Change." Register here.

