The Hour Was Always a Proxy

McKinsey’s clients have started refusing to pay by the hour. The reason they give is blunt: with AI in the room, billing by time spent has become a kind of theater. If the slide deck that took a junior associate three days now takes ninety minutes, what exactly is the client paying for when the invoice still reads “40 hours”? So they’re pushing back. They want to pay for the outcome, or the value, or anything other than the clock.

It is tempting to read this as a story about consultants getting their comeuppance. It is not. It’s a story about what the hour was hiding all along.

The billable hour was never the thing you were buying. It was a proxy. A stand-in for expertise, for judgment earned over years, for the risk a firm absorbs when it puts its name on your decision. Time was just the unit everyone agreed to stop questioning, because it was easy to count and hard to argue with. You couldn’t measure wisdom, so you measured the hours wisdom sat in a chair. The fiction held because nobody had a reason to look underneath it.

AI gave everyone a reason. It didn’t break the pricing model. It revealed that the model was already a story we’d been telling — and that the story and the value had quietly come apart.

The repricing is happening everywhere at once

Once you see it in the consulting invoice, you see it in every headline this week.

Anthropic files to go public, and the coverage frames it as the first real test of AI-boom valuations. But strip away the drama and the question is the same one McKinsey’s clients are asking: are we pricing this by what it earns, or by what we hope it becomes? An IPO is a moment where a private story meets a public ledger. The market gets to decide whether the number people have been repeating to each other in funding rounds survives contact with people who want their money back. It is the billable-hour conversation, scaled to eleven figures.

OpenAI, meanwhile, is rebuilding ChatGPT from the studs — the biggest overhaul since it launched. You don’t rebuild the thing that made you famous because it’s working perfectly. You rebuild it because the ground under it moved, and the version that defined the category three years ago is now just one option among many that do roughly the same thing. The product hasn’t gotten worse. The thing it’s priced against has gotten cheaper. Same repricing, different ledger.

And then there’s Apple, walking into its developer conference with the quiet pressure of a company that built its fortune on polish now being asked to ship intelligence. For decades Apple priced itself on the part you could feel — the weight of the hardware, the smoothness of the glass, the sense that someone cared. That was real value, and it commanded a real premium. But you cannot sand-and-polish your way to a Siri that actually understands you. The market has started pricing a different thing, and a company that spent forty years perfecting the old thing has to decide what it’s actually selling.

The amplifier has a power cord

Underneath all of it sits the least glamorous signal of the week: the fight over data centers. The compute that makes every one of these stories possible has to live somewhere, draw power from somewhere, and increasingly the somewhere is pushing back. Whatever you believe about who’s funding the resistance, the physical fact is simple. The intelligence everyone is repricing runs on electricity, land, and water that someone in a town meeting has opinions about.

This is the part the valuations keep forgetting. Computing is a genuine amplifier — it makes individual people capable of work that used to take institutions, and that’s not hype, it’s the actual engine of the last twenty years of progress. But an amplifier is not free. It has a power cord, and the cord runs back to a grid, and the grid runs back to a county that gets a vote. The cheapest input in the model on the slide is the one that turns out to have a body.

That’s the thread running through every story this week. Cheap to produce, expensive to value correctly. The two move in opposite directions, and AI is the force pulling them apart.

When you build fast feedback into any system, the great gift is that you see the effect of your choices immediately — no lag, no place to hide a bad assumption. That is exactly what is happening to the whole economy right now, only nobody asked for the feedback loop and nobody can turn it off. The cost of making things has collapsed, and the collapse is showing us, in real time, every place where we’d been pricing the proxy instead of the thing.

The consultants felt it first because their proxy was the most naked: time, billed by the hour, in a job that was never really about time. But the hour was always a proxy. So was the valuation, so was the polish, so was the assumption that the power would just be there. We are about to find out, all at once, which of our prices were measuring value and which were measuring a habit. The bill for the difference is already in the mail.