Your AI Strategy Might Be a Slide, Not a Plan

75% of executives admit their own company's AI strategy is more for show than actual guidance. Not a competitor's strategy. Their own. That number comes from a Writer and Workplace Intelligence survey of 2,400 people, published in April 2026. It means three out of four leaders know the plan on the wall isn't the plan running the building.

I sat through a version of this years before AI made it fashionable. A new strategy deck lands. Everyone in the room nods. The meeting ends. Everyone goes back to doing exactly what they were doing before the deck existed. The slide gets filed. Nothing about how work actually happens changes. What's different now is that leaders are willing to say so out loud, in a survey, with their names attached to the answer.

The same survey found 48% of executives call their company's AI adoption a massive disappointment, and 39% still have no formal plan to turn AI tools into revenue. Put those two numbers next to the 75% admitting the strategy is theater, and a pattern comes into focus. Companies aren't failing to write AI strategies. They're writing strategies that were never built to survive contact with a real budget meeting.

I think the tell is where the stress actually lands. 73% of CEOs in this survey report real stress or anxiety about their AI strategy, and 38% describe it as high or crippling. A hard problem doesn't usually produce that kind of anxiety. Knowing your own plan won't survive a direct question in front of the board does.

Meanwhile, 97% of executives say they deployed AI agents in the past year. Adoption and strategy have split apart from each other. Companies move fast on the tools. They move slow, or not at all, on the plan for what those tools are supposed to produce. 59% of companies now spend over a million dollars a year on AI. Only 29% see significant return from generative AI. Only 23% see it from AI agents specifically. The spending outran the plan months ago. Most companies haven't caught up yet.

Here's where it gets uncomfortable for leadership specifically, not just for the org chart below them. 92% of C-suite leaders in the survey say they're cultivating an "AI elite" of top performers. 60% plan layoffs targeting employees who don't adopt AI fast enough. A company that hasn't settled its own strategy still sorts its people into winners and losers based on that unsettled strategy. Employees feel that unfairness before any executive names it out loud. 29% already admit to working against their own company's AI push without saying so. That number climbs to 44% among Gen Z employees specifically.

I keep thinking about how factories handled electricity a century ago. The wiring arrived years before anyone redesigned the plant floor around it. Motors ran. The layout still assumed steam power, belts, and one central drive shaft. Adoption came first. The real plan came only once someone admitted the wiring alone wasn't the strategy. Enterprise AI is sitting in that same gap today. The tools are installed. The layout is untouched.

The relevance cliff explains why this gap costs more than a bad quarter. Every month a company runs on adoption without a real plan, the skills gap between its AI elite and everyone else widens. That gap eventually shows up as attrition, quiet resistance, or a stalled rollout nobody can explain in the next board meeting.

A real AI strategy doesn't need more slides. It needs three plain answers that survive a direct question. Which specific workflow changes because of this, named exactly, not described in general terms? Who signs their name to the result if it doesn't work? And what happens to the people whose current skills this makes obsolete? If your last AI strategy meeting couldn't answer those three out loud, you already know which side of the 75% you're on.

Before your company adds one more agent or one more tool, ask the harder question first. Is this strategy something we'd defend line by line in front of the board, or is it a slide we're hoping nobody reads too closely? A strategy built to survive that question looks different from one built to survive a single meeting, and most companies still have the second kind.

Sources: Writer and Workplace Intelligence, "Elephant in the Office" survey, April 2026 (2,400 respondents: 1,200 C-suite executives and 1,200 non-technical employees).