AI didn't shrink operations — it made operations the surface where growth actually happens.

The loudest takes said ops would go first. The data says otherwise.

For two years the loudest voices in my feed have been running the same script: operations is the low-hanging fruit, the first function AI comes for, the obvious place to cut on the way to an “AI-first” org. Founders internalized it. Some acted on it.

The data coming in now tells a different story — and if you work in operations, it's the story you want to read.

The biggest emerging category of work isn't more engineers to build the AI. It's the roles that take the raw capability and turn it into leverage across the business — AI-augmented operations and frontline work. The people who keep companies running aren't being deleted by this wave. They're becoming the ones who decide whether it produces anything.

That tracks with exactly what I see on the ground.

AI doesn't erase an operator's job. It removes the low-judgment 60% of it — the reporting, the status-chasing, the copy-paste coordination that was never the point — and hands that time back. What's left is the high-judgment work that always was the point: the calls that require context, the tradeoffs that require reading the room, the decisions that require knowing your company and your people well enough to see around a corner. AI is genuinely bad at that part. It's the part operators were hired for and rarely had time to do.

So the function doesn't shrink. It concentrates. It gets more strategic, not less necessary. The operator who used to spend Tuesday assembling a report now spends Tuesday deciding what the report means and what the company should do about it. That's not a diminished role. That's the role, finally unburdened.

Which is why the “cut ops first” instinct is so expensive. We've had the cost-center framing backwards for years. Operations was never the place to trim on the way up — it's the surface on which everything else scales. Sales, product, hiring, compliance: they all move through the operating layer. Weaken it and you don't get lean, you get a company that can't absorb its own growth. Strengthen it, and every other investment compounds.

AI didn't change that math. It sharpened it. By stripping out the mechanical work, AI made the operator's remaining time more valuable, not less — which means the leverage of a strong operations function just went up, not down.

If you're still treating operations as overhead in 2026, I'd gently suggest you're not being disciplined. You're capping your own growth and calling it efficiency.

The companies that figure this out will do the opposite of cutting. They'll put their best judgment — human and augmented — closest to the surface where growth actually happens. That surface is operations. It always was. AI just made it impossible to keep ignoring.