News
What's happening in robotics and AI — curated by the wujec.ai editors.
Six in ten humanoids shipped this year went to a stage or a data lab — not to a factory
Humanoid robots are shipping in numbers that finally look industrial: more than 22,000 units in the first half of 2026, close to 300% up year on year, according to Counterpoint Research. What those machines are bought for is a different story. Two categories still account for more than 60% of all shipments: entertainment and performance, and data production and research — that is, robots bought to appear in front of an audience, and robots bought to generate training data for the models that will one day operate them. Their combined share slipped moderately against last year, but it did not stop being the majority. The segments usually named as the point of the whole exercise remain small. Intelligent manufacturing rose to 13% of shipments; warehousing and logistics to 5%. The figures match what this catalogue has spent August writing about. The two humanoids that broke the 100 m record in Beijing were built to compete, not to work. EngineAI's T800 is the only machine admitted to a humanoid fight league in which every entrant is given identical hardware. DroidUp's Rena performs kunqu opera. None of that is a criticism of the machines — but it is a long way from the images of humanoids carrying parts down a production line that accompany most coverage of the sector. Counterpoint expects the balance to reverse, with service and industrial applications overtaking performance and data collection over the next five years, and total shipments passing 50,000 units in 2026. For now, the most common job for a new humanoid is to be watched.
Humanoid shipments nearly quadrupled in half a year — and every Western maker combined stayed behind China's fifth-largest
Smart Analytics Global published its half-yearly count of humanoid robot shipments on 10 August 2026, and the headline number travelled fast: 19,100 units delivered worldwide in the first six months of the year, against 5,100 in the same period of 2025 — a rise of 272 percent. The second number to travel was the change at the top. AgiBot shipped 8,400 units, a 562-percent jump, taking 44 percent of the market and pushing Unitree, on 5,900 units and 31 percent, into second place for the first time. The figure nobody quoted is the one at the other end of the table. The five vendors the report names — AgiBot, Unitree, Galbot with roughly 900 units, UBTech with about 700 and Leju with about 600 — account for some 16,500 of the 19,100 machines. Chinese vendors, the report says, contributed more than 97 percent of global shipments. That leaves everyone else on the planet, taken together, with fewer than 600 humanoids in six months: less than Leju alone, the fifth-largest producer in a single country. That comparison is worth making carefully, because the milestones Western manufacturers announce are counted differently. Figure said on 23 July that it had built its 1,000th Figure 03 — a cumulative total since production began, not a half-year figure. AgiBot's widely repeated "15,000 robots", announced in June, is cumulative too. The 8,400 and the 5,900 above are six months of deliveries, and mixing the two kinds of number is what makes the gap look smaller than it is. The forecast attached to the report sets up the test for the rest of the year. Smart Analytics Global expects global shipments to approach 60,000 units in 2026, worth roughly 1.6 billion dollars. If the first half really delivered 19,100, the second half has to deliver about 40,900 — a little over twice as much, in five months rather than six. The revenue line implies something else that rarely appears in coverage of this industry: about 27,000 dollars of market revenue per machine shipped. The humanoid business is growing on cheap units, not on the six-figure prototypes that fill the news. One structural shift in the data deserves more attention than the league table. Industrial and commercial deployments accounted for over 70 percent of first-half shipments, up from roughly 50 percent a year earlier, and China absorbed more than 85 percent of global demand. The machines are increasingly bought to work rather than to be demonstrated — and, for now, almost entirely bought at home by the companies that build them. Editorial note: these are the estimates of a private research firm, not audited figures. Manufacturers do not publish delivery numbers on a comparable basis, and the totals should be read as an order of magnitude rather than a count.
AgiBot A3 Ultra →