While everyone is talking about AI, very few are talking about humanoid robots. And almost no one is talking about what humanoid robots are going to do to the price of everything.
That's what I want to get at in this post — the deflationary effect humanoids are going to have on the cost of the stuff we buy every day.
Quick definition: deflationary just means cheaper. Same product, same service, lower price. The opposite of what we've all been complaining about for the last five years.
Humanoid robots — the bipedal kind that walk around and do physical labor — are the easiest version of this story to picture. If they can fold a fitted sheet, they can grind beans, tamp a portafilter, and pull a shot. The capability curve on these things is steep, and the people building them are pouring tens of billions of dollars into closing whatever gaps still exist today.
So with that as the premise, let's use a cup of coffee as the example — because everyone buys one.
## Start at the counter
The median price of a regular cup of coffee on US restaurant menus was $3.65 as of February 2026. In Denver, where I live, a craft latte is more like $5–$6. Let's call it $5 even for the math.
For a typical specialty coffee shop, the rough breakdown looks like this:
- Materials (beans, milk, cup, lid, sleeve): about 20–30% of the price
- Labor (the barista, plus payroll taxes and benefits): 25–35% of the price
- Rent, utilities, equipment, insurance, marketing, fees: another 25–30%
- Profit: maybe 10–15% if the shop is well-run
On a $5 latte, that's roughly $1.25 in materials, $1.50 in labor, $1.50 in overhead, and $0.75 in profit.
Denver just raised its minimum wage to $19.29/hour as of January 1, 2026. With payroll taxes, workers' comp, and benefits loaded on top, the actual cost to the shop is closer to $24–$26/hour for a barista.
## Replace the barista
Tesla is targeting $20,000–$30,000 for Optimus at scale. Unitree's G1 already ships at around $13,500. Industry estimates put maintenance at 10–20% of purchase price per year.
Let's use round, conservative numbers. Buy a humanoid for $30,000. Run it 16 hours a day, 365 days a year — because unlike a barista, it doesn't need to sleep, doesn't get sick, and doesn't call out on a powder day. Add $5,000/year in maintenance and electricity.
Amortize the robot over five years:
- $6,000/year in depreciation
- $5,000/year in upkeep
- ~5,840 working hours per year
That's roughly $1.88/hour of all-in robot cost, versus ~$25/hour fully-loaded for the barista. Call it a 90%+ reduction in the labor line.
Apply that to our $5 latte. The $1.50 of labor drops to roughly $0.11. If everything else stays the same, the latte goes from $5 to about $3.61 — and the shop still makes the same margin it does today. That alone is a 28% price drop. And we haven't touched the supply chain yet.
## Now go upstream
The beans showed up at that coffee shop on a truck. The average truck driver in Denver makes about $26/hour, fully loaded closer to $35–$40/hour. Replace that with a self-driving truck and the per-mile cost drops dramatically.
Before that truck, the beans were at a roaster. A human ran the roaster, packed the bags, drove the forklift. Replace that labor with humanoids running 24/7 and roasting cost drops again. Before the roaster, the beans were on a ship from Colombia or Ethiopia. Before the ship, a processing facility. Every human input in that chain gets compressed by roughly the same 80–90%.
## How low does it go?
If we run humanoids through every layer — picking, processing, shipping, roasting, trucking, brewing, serving — you're looking at a cup of coffee that costs the shop something like $0.75 to deliver. Sell it for $1.50 and the shop still makes a healthy margin.
A $5 cup of coffee becomes a $1.50 cup of coffee. Maybe less. I'm not predicting a date. I'm just doing the math.
## "The savings won't reach the consumer"
This is the objection I always hear. The cynic says: the businesses will just pocket the difference and charge the same $5.
Maybe. In the short term. But this isn't how markets actually work over time. Where there's margin, there's competition. Cafe A keeps charging $5 and pockets a 70% margin. Cafe B opens across the street, charges $3, makes a 50% margin, and steals every customer. Then a humanoid-run drive-thru shows up at $1.50 and crushes everyone else.
Margin attracts competition. Competition compresses price. The savings end up with the consumer. It might take five years, it might take ten — but it gets there.
## Coffee is just the example
Run the same logic on anything. Here's the back-of-the-napkin math on a few things people actually buy:
- Building a single-family home: $428,000 today → roughly $180,000
- House cleaning (standard visit): $180 today → roughly $20
- 30-minute dog walk: $21 today → roughly $2–$3
- Men's haircut: $41 today → roughly $12
- Restaurant burger: $15 today → roughly $6
- Primary care doctor visit (uninsured): $170 today → roughly $50
- Weekly groceries for a family of four: $275 today → roughly $125
- Pair of jeans: $55 today → roughly $18
Every one of these prices is going down, and most of them are going down a lot.
This is the part of the AI conversation almost nobody is having. Everyone's arguing about whether their job is safe. Almost no one is asking the much more interesting question: what does life look like when the cost of producing almost everything drops by 70–90%?
I don't know the full answer. But I know it starts with a cheaper cup of coffee. And I know that if you're building a business right now, the prices you're modeling out for the next ten years are probably wrong — because the cost structure of your competitors, your suppliers, and your customers is about to be rewritten from the ground up.
That's worth thinking about before the next cup.