There’s a familiar frustration playing out in boardrooms across the country right now. Executives are convinced AI will transform their operations. They invest in it, talk about it constantly, position it as central to the company’s future. Then they look down the org chart and realize almost nobody else got the memo.
The numbers tell the story. According to Slingshot’s Digital Work Trends Report, 86% of C-suite executives believe AI usage is required in their company operations. But only 49% of middle managers are actually reinforcing that expectation with their teams. That’s not a gap. That’s a chasm.
The result? For most employees, AI in business still feels optional. Disconnected from how they’re actually evaluated. Something executives care about, not something they need to care about.
This isn’t a technology problem. It’s an execution problem. And it reveals something uncomfortable: having the right tools doesn’t matter if nobody knows how to use them or why they should bother.
The Middle Manager Bottleneck
Here’s the trap middle managers find themselves in. They’re already drowning. The last thing they need is another mandate from above, another tool to learn, another responsibility added to their plate. And if they don’t see immediate results? Why would they invest the energy to teach their teams?
The frustration runs both ways. Employees don’t necessarily understand why they need AI if they’ve been doing their jobs fine without it. Only 2% of employees believe they can’t function without AI. Nobody’s banking on it being a prerequisite yet. So adoption stalls. Nobody’s using it because nobody’s been told what problem it actually solves for them.
But here’s what managers and employees don’t realize: AI doesn’t work like a light switch. You don’t flip it on and suddenly get 30% more productive. It needs time to be trained on real industry expertise. It needs to work alongside human judgment, not replace it.
The fix isn’t complicated, but it does require effort from the top. Executives need to invest in training middle managers specifically, with real examples from their roles and teams. Managers need to understand not just how to use AI themselves, but how to coach their people through integrating it into daily routines. That means spelling out which tasks AI should handle, how to train it properly beyond generic prompts, and how it connects to actual performance metrics.
When that foundation exists, something shifts. Employees gain confidence. Adoption stops feeling like compliance and starts happening naturally.
The Data Problem Nobody Wants to Admit
There’s another layer to this that most organizations gloss over: data literacy.
AI is only as good as the information it learns from. Yet 70% of executives think their employees are constantly using data to make decisions, while only 31% of employees actually say they do. Most people still lean on gut instinct (29%) or wait for someone else to run the numbers (27%).
Part of it is skills. But the real issue is usually simpler and messier. Data exists all over the place. It’s unstructured. Spread across different systems. Sometimes poorly documented. Employees don’t even know what data exists, let alone how to feed it to AI tools.
Organizations that actually want AI adoption to stick need to make data literacy foundational. Show people what data is available, where it lives, which datasets AI actually needs access to. Better yet, connect training directly to real workflows. Show someone concretely how AI can automatically summarize project timelines to identify resource bottlenecks, and suddenly the abstract concept becomes tangible. They see the benefit. They learn by doing.
The Fear Nobody’s Discussing
There’s one more thing happening that leaders aren’t addressing head-on: fear.
Even younger employees, the ones who supposedly embrace new technology, are worried AI will replace them. About 19% of Gen Z workers and 17% of millennials see AI as a competitive threat rather than a tool. That’s not paranoia. That’s a reasonable response to mixed signals from leadership.
When executives talk about AI as a “digital teammate” but never clarify what AI should actually handle versus what humans own, people naturally assume the worst. If you’re not spelling out the boundaries, employees will fill in the gaps themselves. And they’ll usually imagine the scenario that threatens them most.
Leaders need to be explicit. AI handles analysis. AI identifies patterns in data. Humans handle strategy. Humans handle creative decisions that require judgment. Humans decide what comes next. Make those distinctions clear, then actually model them. Normalize conversations about successes and failures with AI. Highlight moments where human judgment saved the day.
The gap between executive ambition and employee reality closes when strategy isn’t just announced from above, but woven into how work actually gets done. When middle managers have the tools and clarity to translate vision into action. When employees understand not just what they’re supposed to do, but why it matters.
Until then, no mandate will stick. The AI transformation that executives imagine will remain exactly that: something imagined in boardrooms, never quite making it to the floors where work actually happens.
