The Hotel Worker Inside Your AI Decision

Every hotel is about to use AI to spend less on labor. I have stood behind that desk, and I need to say what the numbers are leaving out.

I want to start by being fair to the hard-nosed view, because I have lived it from the operations side and I know it is not villainy. It is the job.

Labor is the highest controllable cost in a hotel, and right now it is under real strain. US hotels are on track to pay around 131 billion dollars in wages this year; revenue growth has slowed to where rate increases alone no longer restore margins, and the old comfort, where a good year on the top line quietly fixed the bottom line, is gone. Owners are right to focus on efficiency. Operators are right to be held accountable for it. I have sat on the operations side of a tight budget, and I understand that in a slow-growth year, protecting margin is not greed. It is survival, and disciplined operators have kept hotels open that would otherwise have gone dark.

So, I am not here to tell hotels to stop caring about labor cost. I am here because I spent years behind the front desk, and I know something the spreadsheet does not show. There is a person inside that labor cost, and AI is about to make how we treat that person either a great deal better or a great deal worse. I do not think we are choosing on purpose, and I care too much about this industry to watch us choose badly by default.

I have been the number on that line.

When I read the industry conversation about labor, I recognize the language for one side of it and almost none for the other. We track cost per occupied room to two decimals. We celebrate productivity gains. We benchmark deployment. And we talk about all of it as if the labor were a substance to optimize, rather than the housekeeper, the front desk agent, the night manager who is trying to make the math of their own life work.

I remember math personally. I worked reservations, the front desk, housekeeping supervision, and manager on duty. I know what it feels like to be the cost the hotel is trying to reduce while the world keeps getting more expensive to live in. That is the tension no CPOR chart captures. The hotel is under pressure to spend less on its people at the exact moment those people are watching rent, groceries, and everything else climb faster than their pay. Both pressures are landing on the same human being, pulling in opposite directions, and that human being is the one actually delivering your guest experience.

I am not raising this to moralize. I am raising it because forgetting the person inside the number is not only unkind. It is bad business, and AI is about to make that mistake far more expensive.

AI walks straight into that squeeze.

Into this exact tension, rising labor pressure on one side and rising cost of living on the other, AI now arrives. And it can go one of two ways. The path a hotel takes is really a choice about which person is willing to squeeze.

The first path is the obvious one, and it is where the cost-only mindset leads on its own. Use AI to cut. Remove roles, reduce headcount, and ask whoever is left to absorb the difference at the same pay while the machine picks up the slack. On the spreadsheet, it looks like a win. Labor cost drops, productivity climbs, margin improves. And for a quarter or two, it holds.

Then the real bill arrives, off the spreadsheet, and I have watched this happen. The people who remain, doing more for the same money while their own costs rise, wear out and leave. Turnover, already brutally expensive in hiring, training, and lost knowledge, speeds up. And the guest experience quietly falls apart, because hospitality is delivered by how present and cared-for your people are, and no technology on earth makes an exhausted, resentful, half-staffed team deliver warmth. You optimized the labor line and gutted the product. That is the path where AI deepens the squeeze on the worker first and the business right after, and it is the default unless a leader consciously chooses otherwise.

The second path uses the very same technology toward the opposite end, and it is the one I believe in with everything I have learned in this industry.

Give people their best work back.

Every hotel is full of work no human should ever have done. The manual reconciling between systems that do not talk. The retyping of guest details from one screen to another. The after-hours questions piling up in voicemail nobody can reach. The reports are built by hand. The schedule is made from habit instead of demand. That is the drudgery that eats a hotel worker's day and buries the thing that makes them valuable: their judgment and their presence with a guest.

Used well, AI takes that work away. Not the person. The work. And when you remove the drudgery instead of the human, the math changes. The free time does not get pocketed as a headcount cut. It flows back into the work only a person can do, the real welcome, the anticipation, the rescue of a bad moment, the small, unexpected thing that turns a stay into a story someone tells for years. I know that work is real, because I have got to do it. I once turned a frightened couple's ruined arrival into the best night of their trip, not with technology, but with attention I had time to give. That attention is what drives loyalty and rate, which means the capacity AI frees pays for itself on the revenue side, not just the cost side.

And here is what the cost-only view cannot see. A hotel that uses AI this way can afford to treat its people better, not worse, because it is getting more value from the same team instead of the same value from a squeezed one. Efficiency and humanity stop being enemies. The technology that lets you run leaner is the same technology that lets you pay and keep good people, because you are capturing more value per person rather than wringing more hours out of them. That is what I mean when I talk about value innovation, and this is the most human place it applies.

Let me be honest about the hard parts.

I will not pretend this dissolves every difficult trade-off, because I respect you too much to hand-wave.

Sometimes the conflict is real. Some roles genuinely do get made redundant, and no gentle framing changes the fact that a specific person may lose a specific job. Pretending AI costs no one anything is exactly dishonesty that makes workers distrust this whole conversation, and they would be right to. The truth is not that AI harms no one. It is that hotels have a genuine choice about whether to use it mainly to cut people or mainly to free them, and that choice has been made by default, toward cutting, without anyone deciding it.

It is also true that the humane path is harder. Cutting a role is simple and shows up next quarter. Redesigning work so AI removes the drudgery and people move up to higher-value contribution takes clarity, patience, and returns that build over time. A hotel in genuine financial distress may not feel it has room for the harder path, and sometimes survival really does force the blunt instrument. I will not lecture an operator staring down a crisis. But to the operator who still has a choice, I will say plainly that the choice matters more than it looks, and that the blunt path is the more expensive one over any horizon longer than a couple of quarters.

And I will own my values here rather than smuggling them in. I believe a hotel owes something to the people inside its labor cost; beyond the bare minimum the market allows. I believe that as someone who was one of those people. But I am also telling you it is a smart position, not only the decent one, because the entire product you sell is delivered by those people, and any business model that quietly runs on burning them out is not efficient. It is borrowing against a future it will repay in turnover, weak reviews, and a guest experience it can no longer deliver.

What I believe is AI.

Let me tell you what AI is to me, because it is not what most of the fear and the hype would have you believe.

AI was built by people, for people. Every bit of what it can do traces back to human knowledge, and in a hotel, that knowledge lives in the people who know the property best. AI is only ever as good as the data it is fed, and the richest data in any hotel is not in a system. It is in the housekeeper who knows which guests return and why, the front desk agent who remembers the anniversary, the team that reads a room before a word is spoken. That knowledge is the fuel. Which means the people are not the cost that AI reduces. They are the source of everything AI can give a hotel in the first place.

That changes what this technology is for. To me, AI is a way for employees to take ownership of the whole hotel ecosystem, to see how their piece connects to every other piece, and to collaborate on finding new ways to create benefit for everyone: the guest, the hotel, the owner, and themselves. It is a system that finally lets us reward each other, because when the frontline's knowledge is captured and honored rather than lost, everyone shares what it produces.

That is why I keep saying AI amplifies people, and never the other way around. The moment we invert that, the moment we treat people as the thing to be reduced and AI as the thing that matters, we have not just made an ethical mistake. We have misunderstood what we are holding. AI is an amplifier. Point it at people who are trusted, developed, and given ownership, and it makes them extraordinary. Point it at people you are trying to eliminate, and it amplifies nothing, because you have thrown away the very source it runs on.

The choice underneath the number

So here is where I land, having given the cost side its fair due.

The labor pressure is real, and the discipline it demands is legitimate. But we are having an industry conversation about labor cost that has forgotten there is a person inside every unit of it, someone being asked to accept less from the hotel at the very moment the world is charging them more to live. AI is about to drag that forgotten tension into the light, because it hands every hotel a lever strong enough to make the squeeze much worse or genuinely better, and most hotels are grabbing that lever without deciding which they mean to do.

This was never really a technology decision. It is a decision about what we believe a hotel owes the people who are the hotel. Use AI to remove people and squeeze the rest, and you will win a few quarters and lose the thing you are selling. Use it to remove the work that wasted them and give that time back to what only humans can do, and you get the efficiency, the demand, and the humanity this business quietly runs on, both at once.

There is a person inside your labor cost. I know, because I was one. The most important question of this whole AI moment, for a hotel, is whether we use this technology to squeeze that person harder, or to finally give them their best work back. Over any horizon that matters, math rewards the second choice. So does everything that made me fall in love with this industry in the first place.