You’ve felt it. That sinking feeling when your boss announces the company is going all-in on AI, and you know it’s a disaster but you can’t say a word. The room goes quiet. Anyone who questions the plan is branded a Luddite, a dinosaur, a traitor to progress. You nod along, because the alternative—professional excommunication—is too terrifying.
AI mania isn’t a strategy; it’s a religious conversion, and heretics are excommunicated. This is the dirty secret that nobody in the C-suite will admit: the decision to adopt AI has nothing to do with rational analysis. It’s a reflexive feedback loop where industry hype dictates corporate strategy, overriding every ounce of common sense. The louder the hype, the more irrational the bets. And the more irrational the bets, the more silence is demanded from those who see the emperor has no clothes.
A product manager at a Fortune 500 company recently confessed to me: ‘We now generate 10x the code using AI, but our review pipeline is drowning. The real bottleneck is human trust. We can’t verify half of what the AI produces, so we just ship it and pray.’ That’s not productivity. That’s an abdication of responsibility.
The 100x productivity multiplier is a lie unless you’re measuring the wrong thing. The touted ‘100x’ isn’t solving actual business bottlenecks; it merely accelerates the generation of low-value output. The real bottleneck has shifted from production to verification and filtering. But nobody wants to talk about that because the narrative is too seductive. ‘We’re 100x faster!’ sounds great in a board meeting. ‘We’re drowning in untrustworthy garbage’ does not.
This is dangerous. It’s not just wasteful; it’s corrupting decision-making at every level. Leaders who should be asking hard questions about ROI, risk, and organizational capacity are instead asking, ‘How do we catch up with OpenAI?’ They’re not thinking—they’re reacting. And they’re dragging entire teams into a vortex of busywork masked as innovation.
The irony? The very people who are most skeptical are the ones who’ve actually tried to implement AI at scale. They’re the engineers who’ve seen the hallucinations, the product managers who’ve watched quality metrics crater, the executives who’ve realized that faster bullshit is still bullshit. But they’re silenced by the fear of being labeled ‘anti-progress.’
The best AI strategy right now is a moratorium on AI strategy. Before you launch another pilot, before you mandate ‘AI-first’ for every project, stop. Ask yourself: What problem are we actually solving? Whose job is it to verify the output? What happens when the hype dies down and we’re left with a mountain of untested code and broken promises?
Stop nodding along. Start speaking up. The heretics are the only ones who can save us from this cult. And if you’re afraid to be the heretic, at least remember this: history doesn’t remember the ones who stayed silent. It remembers the ones who said, ‘This is madness,’ while everyone else was cheering.
FAQ
Q: Aren't there legitimate use cases where AI provides 100x productivity gains?
A: Yes, in specific narrow domains like code generation for boilerplate or data extraction from unstructured text. But the claim that AI universally multiplies productivity by 100x is marketing, not reality. The bottleneck shifts to verification, and the cost of errors often outweighs the speed gains.
Q: So what should a manager do if their company is pushing AI adoption?
A: First, demand a clear problem statement. What specific bottleneck does AI solve? Second, allocate equal resources to verification and quality assurance. Third, establish a 'red team' that can safely dissent. If you can't honestly question the strategy, the strategy is already broken.
Q: Isn't it possible that the skeptics are just afraid of change?
A: Some skeptics are indeed change-averse, but the most vocal critics are often the ones who have actually implemented AI and seen the messy reality. The contrarian take is that blind adoption is more dangerous than thoughtful skepticism. The real failure is not moving fast enough—it's moving without thinking.