Here’s a head-scratcher: Over a third (32%) of employees hide their AI usage from employers. The reasons vary from fears over job security to maintaining a personal productivity advantage. Many don’t want colleagues questioning their capabilities. 

I get it. When more than half of the workers we surveyed believe that being more efficient only leads to more work rather than better work, who can blame them for keeping productivity gains to themselves? 

But this creates another, bigger problem. Without visibility into how people actually work, companies invest in the wrong tools, create ineffective policies and miss the opportunity to capture successful patterns that could benefit everyone. This is a costly problem. In this year alone, Forrester estimates that companies will spend north of $4.9 trillion on technology, including software, IT services, cloud and GenAI tools, a 5.6% increase over last year. However, tech debt costs $2.41 trillion a year in the U.S. alone, and a recent report indicates it would take $1.52 trillion to fix. Forget the ROI. 

How Tech Debt Accumulates 

I’ve seen this play out with CIOs and other execs I’ve spoken to who’ve gone all-in on AI — fancy chatbots, automated help desks, the works. Almost every conversation eventually circles back to a similar frustration, with some variation on this theme: “We’ve spent X dollars, but we’re not seeing the returns we expected.” 

Regardless of industry or specific AI efforts, these frustrations seem to boil down to the same culprit. Their AI initiatives continue to stumble over decades of accumulated tech debt. 

Part of the reason is despite the hype, most organizations use AI — let’s say, timidly. Fewer than half employ it for predictive maintenance or detecting network anomalies. Fewer than a third use it for root-cause analysis or intelligent ticket routing. 

Why such hesitation? Because implementing AI effectively means confronting all the messiness that came before. It means admitting our tech environments need a serious cleanup before adding another layer of complexity. Tech complexity has become a monster.  

This mess came from years of bolting on new systems without retiring old ones. Some IT professionals point to redundant applications as a major source of wasted budget and others blame overprovisioning in the cloud — the digital equivalent of paying rent on empty apartments. 

Meanwhile, 48% of companies still run software that’s reached end of life. Every CIO I know would consider it a total dealbreaker if someone ran Windows XP on the corporate network, yet somehow we’ve normalized keeping zombie applications alive throughout our organizations. 

On top of that, data silos make everything worse. Security and IT data remain trapped in departmental fortresses and this fragmentation wreaks havoc by slowing down security response times and weakening a security posture. IT teams admit something that, to me, is alarming: Their infrastructure has grown so tangled they can no longer maintain basic security practices. Let that sink in. Companies with eight-figure tech budgets can’t reliably patch vulnerable systems or implement fundamental security controls. 

No one builds silos deliberately. Silos emerge from organizational boundaries, competing priorities and the way we fund and manage projects. But AI can’t work its magic when it can only see fragments of the whole picture. 

A Roadmap for Maximizing Tech’s ROI 

Companies making real progress share a few traits: 

• They’ve stopped pretending complexity doesn’t exist. The first step is acknowledging the problem. 

• They track what they own. It sounds basic, but you can’t optimize what you don’t measure. 

• They focus on integration before innovation. New tools must play nicely with existing systems. Period. 

• They’ve broken down departmental silos that keep critical data trapped. This doesn’t happen by accident — it requires deliberate effort and executive sponsorship. 

• They recognize AI as both a solution and a potential problem. Without governance, AI implementations can create new silos and increase sprawl rather than reduce it. 

• They balance optimization with empowerment, using technology both to make operations leaner and to give people greater flexibility in how they work. 

AI will absolutely transform how work gets done. But the transformation won’t look like the glossy vendor decks suggest. The largest AI budgets don’t necessarily translate to the largest benefits. Better ROI will come from a willingness to address years of accumulated tech debt first. Companies that understand AI need a solid foundation of clean data, streamlined infrastructure and collaborative processes to deliver meaningful returns. 

If you’re frustrated by underwhelming results from your AI investments, take a hard look at what’s underneath. Sometimes you need to fix the foundation before adding another floor.