AI doesn't take jobs. Leverage does.

The fear is real, and I understand where it comes from. Every major technology shift in the last hundred years has been followed by a wave of anxiety about jobs disappearing — sometimes justified, sometimes not. AI is the latest version of that fear, and it's amplified by the fact that this particular shift is showing up faster than most of the earlier ones.

But the math of how AI actually changes a business doesn't match the fear. Not in the cases I've seen up close, and not in the ones I've built.

Here's what actually happens. If AI makes a team ten times more productive on a specific task, the business doesn't need one-tenth of the team. It needs more of the team doing different work, because the business can now deliver ten times more output — which means ten times more customers to manage, ten times more quality to check, ten times more edge cases to handle, and ten times more opportunity to pursue. The bottleneck moves. The work moves with it.

None of that is automatic. A business that cuts staff the moment automation lands is a business that was planning to cut staff anyway and found a convenient reason. A business that thinks of AI as leverage — as a way to do more, not a way to do the same with less — ends up hiring more people, not fewer. The roles change shape, but the headcount tends to grow.

The work that goes away is the work that was never really work. Copying invoice line items from paper to a screen. Reformatting a PDF someone sent you. Pulling numbers from one system to paste into another. Nobody went into their career to do that, and nobody who does it well feels valued for it. When AI takes that work away, it doesn't eliminate the person — it frees up the person to do the things the business actually needed them to do but never had time for.

The other half of this is the part people don't say out loud: if AI can really do a job, that job was probably underpaid and overlooked for a long time before AI showed up. The jobs that survive and grow are the ones that require judgment, relationships, context, and the ability to deal with ambiguity. Those jobs get more valuable, not less.

What I'd actually worry about, if I were worrying, is something different. Not that AI takes jobs, but that organizations use AI as a reason to avoid thinking clearly. It's easier to layoff-and-replace than to figure out what the business should be doing with its newfound capacity. That's a failure of management, not a failure of technology. And it's also a choice — one that a lot of businesses won't make, because the ones who use AI as leverage will out-compete the ones who use it as a cost cut.

The jobs that are at risk aren't the ones the alarmists are pointing at. They're the ones where no human judgment is actually being applied — the ones that were already being done on autopilot. Those were always going to go, AI or no AI. Everything else — the judgment, the context, the relationships, the tacit knowledge — that's where the work is now, and where the work will stay.

The fear is understandable. It's just pointed at the wrong target.