AI Is Growing Up: Why Companies Are Hitting Pause and Getting Real About Results

There’s a noticeable shift happening in how companies are talking about AI, and I don’t think it’s being read correctly.

The easy headline is that organizations are pulling back because AI is too expensive and the return isn’t there. That’s true in some cases, but it misses the real story. What’s actually happening is that the conversation is getting more honest.

A year ago, AI felt inevitable. Every leadership team felt pressure to show progress. That meant pilots, vendor deals, and some version of “transformation” in motion. There was real energy and excitement behind it, but not always very much discipline. When things move that fast, the costs don’t show up cleanly. They show up later, and often in places no one modeled very well.  And now those bills are landing.

Not just for software or model access, but for everything around it. Data has to be organized, cleaned, and governed. Systems have to connect. Teams need to be trained. Processes have to be reworked. None of that is small, and none of it is free. And when you actually add it all up, AI stops looking like a feature and starts looking like a core operating decision.

This is where a lot of companies I work with are hitting pause.  Not because they doubt the technology, but because they are trying to make the numbers actually work within their budgets.

From where I sit, both in the classroom and through Enroot, I believe this is a healthy moment. It’s the shift from curiosity to accountability in the Intelligence Age. And we should have expected it. This isn’t the first time we’ve seen this.  The internet went through it, Cloud Computing went through it, Big Data, Mobile Apps, RPA all went through it. Early excitement gives way to scrutiny. It’s uncomfortable, but it’s also where the real value gets built when better questions are asked.

The problem isn’t that AI doesn’t work. It’s that too many early efforts weren’t tied to outcomes in a way that made the return obvious. When something is framed as “transformational,” it’s hard to define what success looks like in the near term. So, you get pilots that are interesting, sometimes impressive, but tough to defend when budgets tighten.

The questions are getting better now. Not “What can AI do?” but “What is it doing for us?” That’s a different conversation that forces prioritization, tradeoffs and clarity around where this shows up in revenue, in cost savings, or in productivity that can actually be measured.  It also puts pressure on how these solutions are priced and delivered.

One thing I keep hearing is that it isn’t just the total cost that frustrates teams. It’s the unpredictability. Usage-based pricing that spikes. Add-ons that weren’t clear upfront. Implementation costs that grow as projects get more complex. Over time, that uncertainty starts to feel like risk, and risk slows adoption faster than almost anything.

That’s something we’ve been deliberate about with EnrichIT!.

If you’re asking a business to bring AI into how it manages and grows customer relationships, you owe them a clear understanding of what it will cost. Not just at the start, but as they scale. Transparent pricing sounds basic, but it’s rare right now. We made the call early that there shouldn’t be surprises. It’s a better experience for clients, and it forces us to build in a way where cost and value stay aligned.

There’s another shift that I find encouraging. Companies are getting more selective about where AI fits. Instead of spreading it everywhere, they’re focusing on use cases where the impact is easier to see and defend. Data enrichment is a good example. If your CRM improves, your targeting improves. If your targeting improves, your pipeline quality changes. Those are connections people understand.  And this kind of clarity is what drives adoption. Not big, abstract stories about transformation, but simple, credible paths to results.

So yes, there is a pullback of sorts. Some projects are being reevaluated. Some budgets are tightening. But I don’t see that as a loss of momentum. I see it as a correction that was always coming.

We’re past the point where saying “we’re doing AI” is enough. Now it must work, and it has to make financial sense. That’s a higher bar, but it’s the right one.

Over time, this is what will separate what lasts from what doesn’t. The companies that get this right won’t be the ones that moved fastest last year. They’ll be the ones that are clearest right now about where AI fits, what it costs, and what it delivers.

That’s a much better place to build from.