Hugh Cumming, CTO, Vena.gettyAI is going to introduce massive change quickly. There’s no guarantee those changes will get us where we want to be.​I’ve been a technology professional across a variety of industries for over 20 years. In that time, I’ve led organizations through many technology changes—ERP rollouts, cloud migrations, Internet 1.0 and mobile. Each one followed a similar pattern: rapid adoption, visible momentum and, eventually, cracks in the foundation.This time feels different because AI is moving faster than any of them. ​McKinsey researchers show that 62% of organizations are already experimenting with AI agents. Many are going further—automating workflows, building custom solutions and deploying agents across the business. That kind of acceleration is unprecedented.​Most businesses see that as a good thing. I do too—mostly. But after two decades of leading through technology transformation, I’ve learned to pay attention to what’s happening underneath the momentum. And right now, a few things concern me.What if AI only creates the illusion of progress?What concerns me isn’t the pace of adoption. It’s how familiar the pattern looks. I’ve seen versions of it before—rapid experimentation, local success stories, growing complexity and a widening gap between activity and measurable business value.There are lessons from past technology cycles that can help. But only if we’re willing to look back before we move forward.Lesson 1: More tools can create more complexity.Before joining Vena, I worked at a large financial services company where IT sat firmly at the center of growth. If the business needed something, we built it. But gradually that model shifted. The company moved capability out of IT and into the hands of the business. The number of applications grew—eventually exceeding 1,500, tracked on a spreadsheet that itself became a problem to maintain.Underneath, a different reality was taking shape. Data was inconsistent and decentralized. Processes diverged. What started as empowerment introduced a layer of complexity that required governance, security and standardization nobody had planned for.That’s the pattern AI is following now. The challenge is that AI outputs look reasonable enough to pass without scrutiny. By the time the inconsistencies surface, they’re already embedded across teams, workflows and decisions. Agents, workflows and bespoke solutions multiply. So do inconsistencies, silos and rework.Lesson 2: Technology only creates value when it changes how work happens.AI doesn’t fix broken systems. If decisions were slow before, AI gives you more options faster—but options aren’t decisions. The right data is still what determines whether you choose correctly, and AI doesn’t fix that problem. You end up moving quicker, potentially in the wrong direction.To get real value from AI, you need to find out what work needs to change, where human judgment matters and what "good" looks like when the constraints are different.At Vena, we built an "AI Champions" program around exactly that kind of thinking. People come together quarterly for intensive sessions, not just to learn the tools but to apply them to real problems in their own work. Finance analysts generate scenario models in minutes. Sales leaders structure discovery calls differently. Marketing managers test messaging variations at a pace that wasn’t possible before. The technology is the same as everyone else’s—what’s different is that people are using it to change how they work, not just how fast they work.Lesson 3: Solving the wrong problem at scale only accelerates waste.Early in my career, I cofounded a click-to-chat platform called WebHelp. We launched as the dot-com bubble was about to burst, building real-time customer support for companies navigating a new digital frontier. The company was acquired, and by most measures, we did well.But while we were building better support interfaces to facilitate ordering, one company was focused somewhere else entirely. Amazon understood that the real problem wasn’t capturing orders—it was fulfilling them. That insight changed the economics of the entire business.​Every major technology wave changed how work was executed. AI may be the first that changes who does it. We’re already seeing it in finance—not analysts doing the same work faster, but analysts doing fundamentally different work. The reconciliation and the manual forecast updates are gone. What’s left is judgment and interpretation—the things that rely on what makes us unique. What It All MeansThe danger of AI isn’t that it will fail. It’s that it will succeed well enough, and for long enough, to look like progress—without ever driving real change.AI is going to introduce massive change quickly. There’s no guarantee those changes will get us where we want to be.Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?