Morgan Stanley has now raised its forecast for China’s humanoid robot shipments three times in six months, which is either a sign that the future is arriving faster than anyone modeled, or a sign that a bank cannot stop revising the same spreadsheet upward. This week it landed on 50,000 units shipping in 2026 — nearly double the 28,000 it had projected only weeks earlier, which was itself double the 14,000 it forecast back in January.
A forecast that keeps chasing itself
The arithmetic is worth sitting with. January: 14,000. Spring: 28,000. Now: 50,000. The bank has more than tripled its own estimate inside a single calendar year, and the year is not over, per the South China Morning Post and CNBC. Analyst Sheng Zhong attributes the acceleration to “commercial verification, policy support, and supply-chain feedback” — China is, in his read, moving from demos to deployments faster than the models assumed.
The longer-range numbers are the eye-catching ones. Morgan Stanley pegs China’s humanoid market at roughly $2 billion this year, growing to $15 billion by 2030, with annual shipments reaching 446,000 units by the end of the decade. Those are the figures that get screenshotted and pasted into pitch decks. They are also, like all 2030 figures, a story told with a straight face about a year nobody can see yet.
Shipments are not the same as buyers
Here is the part that deserves a raised eyebrow. Everything in this forecast measures supply — how many humanoids will roll off Chinese lines and into the channel. None of it directly measures demand, meaning how many of those units will be bought, installed, and kept working by an end customer who renews. That distinction is not pedantic. It is the entire ballgame.
China’s humanoid surge is being pushed hard from the supply side: policy support, provincial subsidies, an actuator and component base that no other country can match on cost, and a fleet of manufacturers racing to hit volume targets. Build 50,000 robots and, by definition, 50,000 robots ship. Whether 50,000 factories, warehouses, and service businesses actually want them at a price that makes the unit economics work is a separate question — and it is the one the forecast quietly steps around. CNBC’s own China desk spent part of June asking the obvious follow-up out loud: who, exactly, is going to buy all these humanoids?
Why this matters for the displacement story
For readers tracking what automation does to actual jobs, a doubled shipment forecast is not the same as doubled displacement. A robot sitting in a distributor’s warehouse, or running a subsidized demo at a provincial showcase, has not replaced anyone. The labor impact only lands when these units are deployed into real work at a cost lower than the human alternative and kept running. The supply curve and the displacement curve are related, but they are not the same line, and right now the supply curve is the one Wall Street keeps redrawing.
So take the 50,000 for what it is: a genuine signal that China has decided to flood the zone with humanoids, backed by real manufacturing capacity and real state will. That part is not hype. What remains unproven is the demand underneath it — and a forecast that has tripled in six months while still measuring the easy variable is worth reading as ambition, not yet as arrival. The robots are coming off the line. The harder question is who keeps them switched on after the subsidy ends.
