In 1991, Geoffrey Moore wrote a book that changed how Silicon Valley thought about technology markets. His insight was deceptively simple: there is a massive psychological gap between the people who eagerly adopt a new technology and the people who will eventually use it at scale. He called it the chasm. Most companies, he argued, fall into it. What Moore couldn't have anticipated is that thirty-five years later, the most important chasm companies need to cross isn't in their market. It's inside their own walls — between the employees who have embraced AI and the ones who haven't.
Moore's original model describes five types of technology adopters arranged along a bell curve. First come the Innovators — the 2.5% who chase new technology for the joy of it. Then the Early Adopters — visionaries who see competitive advantage where others see risk. Together they represent the early market. But the far larger population — the Early Majority, the Late Majority, the Laggards — lives by different rules. They need proof. They need workflow fit. They need to see someone they trust succeed before they'll budge. In the enterprise AI moment, these same psychographic archetypes are playing out not across customer segments but across office floors. The bell curve is your org chart.
The AI Vanguard (≈3%) are Moore's Innovators — the people so far ahead of the curve that they've stopped thinking about it as a curve. They're not just using AI; they're constructing with it. Building internal tools, designing new workflows, and automating entire job functions before anyone asked them to. They're the ones who will brief the CEO on what's possible, get frustrated that adoption is slow, and occasionally forget that not everyone thinks the way they do. They represent a small fraction of the workforce, but their fingerprints are on every serious AI initiative inside the organization.
The AI Explorers (≈12%) are Moore's Early Adopters — and they're already across. Saving nine hours a week. Getting promoted at three times the rate of their peers. Quietly becoming indispensable, without waiting for permission or a mandate. Together with the Vanguard, they make up roughly 15% of your workforce. The remaining 85% are somewhere else on the curve: watching, waiting, or actively resisting. Recent data makes this stark: 85% of employees have access to AI tools, but only 25% use them regularly. The tools crossed the chasm. The people haven't.
The AI Pragmatists (≈34%) are your largest immediate opportunity, and the most misread group in the organization. They are not resistant. They are waiting for proof. Specifically: a colleague who has won with this, a workflow that actually fits how they work, and a use case they can point to when someone asks why they changed. These are competent professionals who aren't opposed to disruption — they're opposed to disruption without a demonstrated payoff. Moore had a term for what they need: the "whole product." Not just a ChatGPT license, but the complete ecosystem that makes a technology viable for mainstream use — integrated workflows, visible wins, training that maps to the actual job. Give them that, and they move. Without it, they stall. And when they stall, they take a third of your organization with them.
The AI Skeptics (≈34%) are watching closely, and they are not convinced the organization will manage this fairly. This is a reasonable position. Sixty percent of C-suite executives admit to cultivating "AI elite" employees while planning reductions for non-adopters. The skeptics see this. Their caution isn't paranoia — it's pattern recognition. They're not against AI; they're against the version of AI adoption that treats them as a cost to be optimized rather than a professional to be developed. Moore understood that the mainstream market doesn't just need a better product; it needs trust in the category itself. For the AI Skeptics, rebuilding that trust is a management problem before it is a technology problem.
The AI Resistors (≈16%) are the smallest group but the loudest signal. In 2026, Fortune reported that 80% of white-collar workers are quietly rebelling against AI mandates. This isn't Luddism. It's what happens when technology is done to people rather than with them — when the rollout is a directive, not a conversation; when the benefits flow to shareholders and the risk lands on workers. The resistors aren't wrong that something is being mishandled. They're just expressing it in the only way that feels available to them.
Here is what Moore's framework ultimately teaches about this moment: you cannot drag people across the chasm with mandates. You cross it by finding a beachhead — a specific team, a specific use case, a specific workflow where AI demonstrably wins — and letting that success story do the work. The Vanguard are already so far across they've forgotten there was a gap. The Explorers are right behind them. The Pragmatists need a colleague who's made it and a path they can follow. The Skeptics need to see the organization handle this with some integrity before they'll move. The Resistors need to feel like participants, not targets. Each group requires something different — and mandates reach none of them. The companies that get this right won't just be more efficient; they'll have built an organizational capability that compounds over time. The ones that don't will find that the chasm runs right through the middle of their workforce — and spend years wondering why the technology never delivered.
