We surveyed 123 credit union executives across functions—CEOs, finance leaders, IT, lending, operations—to get a clearer read on AI readiness inside real organizations.
Three numbers are worth putting next to each other:
- 3.97 / 5 — personal comfort discussing AI
- 3.02 / 5 — perceived leadership team knowledge
- 2.40–3.41 / 5 — departmental readiness
They’re all coming from the same survey. They don’t describe the same reality.
Personal comfort is high enough to feel fine
Start with the easiest one.
A 3.97 out of 5 average for personal comfort means most executives aren’t intimidated by AI anymore. They’ve read enough, heard enough, maybe played around with tools a bit. Conversations aren’t awkward.
That number tells you people don’t feel behind individually.
And to be fair, they’re not wrong. Compared to where things were even a year ago, the baseline level of familiarity has moved up pretty quickly.
Team knowledge is where it starts to wobble
Drop down one level and the tone changes.
Leadership team knowledge sits at 3.02 out of 5.
That’s not a disaster, but it’s not especially strong either. Most responses cluster in the middle, which usually means the same thing in practice: some people get it, some people don’t, and everyone is kind of working around that.
This is where things start to get a little messy. AI stops being an individual topic and becomes a group problem.
You can still move forward from here, but it takes more coordination than most teams expect going in.
Departmental readiness is where the gap shows up
Once you move out of the leadership layer and into actual departments, the numbers spread out.
On the higher end:
- Marketing & Member Experience: 3.41 / 5
- Risk & Compliance: 3.29 / 5
Closer to the middle:
- Technology / IT: 3.13 / 5
- Lending & Credit: 3.07 / 5
- HR: 3.00 / 5
And then there’s Finance:
- 2.40 / 5
That range—roughly 2.4 to 3.4—is doing a lot of work.
It shows you that “the organization” isn’t a single thing when it comes to AI readiness. Some teams are experimenting. Some are cautiously watching. Some are still trying to figure out where this even fits.
Finance being at the bottom isn’t surprising, but it is important. That’s the function that ends up pressure-testing every initiative, whether it’s formally part of the process or not.
Put the three together and it gets clearer
On their own, none of these numbers are that dramatic.
Together, they tell a pretty specific story.
Executives feel comfortable talking about AI.
They’re less confident in how much their peers actually understand it.
And when it comes to doing something with it across the organization, readiness is uneven enough to slow things down.
Not because anyone is actively resisting it, but because the organization isn’t moving at the same speed everywhere.
This is where most AI conversations get misleading
A lot of AI discussion—especially at conferences or in industry content—happens at that first level. Personal comfort, general awareness, maybe a few examples of what’s possible.
At that level, everything feels like it’s moving quickly.
Once you step into an actual organization, you’re dealing with different levels of understanding, different incentives, and different levels of readiness across teams. That’s where things slow down.
The gap between those two views is where most frustration comes from. From the outside, it looks like progress should be faster than it is.
Why this matters more than it looks
If personal comfort and organizational readiness were aligned, most of this wouldn’t be an issue.
The problem is that they’re not.
People who feel comfortable with AI tend to assume the organization is closer to ready than it actually is. That leads to overestimating how quickly initiatives can move, or underestimating how much coordination is required to get something off the ground.
Meanwhile, teams that are less prepared end up reacting instead of participating, which creates even more uneven progress.
What to do with this
The takeaway here isn’t that credit unions are “behind” on AI.
It’s that readiness isn’t uniform, and treating it like it is creates problems.
The more useful approach is to recognize the three layers separately:
- Individual understanding
- Leadership alignment and shared knowledge
- Department-level readiness
Each one needs to be addressed on its own, because strength in one doesn’t automatically carry over to the others.
Most organizations are already in a decent place on the first layer. The second is mixed. The third is where most of the work still is.
Until those three start to line up, progress is going to feel slower than the conversation around AI suggests it should be.
Want more insights like these delivered directly to your inbox every week?
