Why is China’s humanoid robot industry winning the early market?

Chinese humanoid robots have attracted global attention with kung fu moves at the country’s televised Spring Festival ceremony, while Chinese phone maker Honor is set to unveil its first robot at MWC in Spain.

Robotics has been designated a priority under the country’s “Made in China 2025” plan, although it was originally focused on factory automation, rather than humanoid robots. Now, rapid progress in multi-modal AI is accelerating so-called embodied AI — autonomous machines operating in the real world — a boost that officials say could help offset labor shortages and drive productivity gains.

At this early stage in the development of humanoid robots, Chinese companies are outpacing their American competitors in terms of speed and scale, said Selina Xu, a China and artificial intelligence policy officer in Eric Schmidt’s office.

“China has a more robust hardware supply chain — most of which was built by the electric vehicle sector, from sensors to batteries — and the strongest manufacturing base in the world, allowing companies to iterate much faster than Western competitors,” Xu told TechCrunch.

As a result, Chinese robots are not only cheaper, but companies can also launch new models more quickly, Xu noted, adding that leading Chinese company Unitree shipped about 36 times more units last year than US rivals Figger and Tesla.

Global humanoid robot shipments totaled just 13,317 units last year, according to a Forbes report released last month. That’s a small base for an industry that’s expected to roughly double annually and reach 2.6 million units by 2035. (However, the numbers should be viewed with caution. The report notes that it’s still unclear how many units represent commercial sales versus experimental models or pilot deployments, highlighting the early-stage nature of this industry.)

Leading manufacturers of humanoid robots by 2025 are Chinese companies Agibot and Unitree, followed by UBTech, Leju Robotics, Engine AI, and Fourier Intelligence, underscoring Beijing’s early dominance of the sector.

The biggest recent shift has been from “demo-driven excitement” to “operations-driven adoption,” Yuli Zhao, chief strategy officer at Galbot, told TechCrunch. The humanoid galbot, G1, appeared at this year’s Spring Festival Gala, China’s annual state-run television show on Lunar New Year’s Eve, alongside robots from companies Unitree Robotics, Noetix and MagicLab.

“More customers are asking: Can a robot operate stably in real environments and actually take the work off people’s shoulders? This practical appeal has been strengthened in China because industrial policy and strategy encourage automation upgrades, and the manufacturing ecosystem makes iteration very fast,” Zhao said.

Zhao added that while increased funding for human-powered startups has “certainly accelerated” the pace of progress, “the most lasting adoption comes when you can demonstrate reliable, repeatable value in production or service processes, not just a one-off offering.”

Still, investment is helping, and Chinese robot makers are working to secure it. Last year, Unitree was valued at about $3 billion after closing its Series C, with ambitions to reach as much as $7 billion in a future IPO. Meanwhile, Galbot has raised more than $300 million in new funding, taking its valuation to $3 billion, one of the largest financings in China’s humanoid robotics sector to date.

US companies are also moving beyond flashy demos to focus on real-world deployments. In addition, they pursue their aggressive goals. For example, the American Startup Foundation plans to build 50,000 humanoid robots by the end of 2027.

But China is already targeting a mix of mass-market models and affordable, cutting-edge applications, leading to the rapid expansion of humanoid robots across industrial, consumer and rehabilitation sectors, according to a December TrendForce report.

Bottlenecks facing China’s hegemony

When it comes to AI systems and integrated software, it’s still unclear where Chinese robotics companies really stand. The industry is betting largely on vision, language, action and “universal models,” but both technologies are still in the early stages. Nvidia currently leads the field with its comprehensive human software portfolio, according to Xu, so it’s only natural that most human-powered startups in China will be powered by Nvidia’s Orin chips. However, she said local chipmakers are developing local alternatives.

However, humanoid robot makers are still working on basic problems. The challenge is to enable basic robot models to predict the “next physical state” the robot will encounter in unpredictable environments, much like how large language models predict the next word. But unlike MBAs, humanoid robotics companies can’t simply use the Internet to get training data, Xu says. Most of them therefore rely on simulation environments, which generate synthetic data, although real-world data collection remains necessary.

“Because of the problem of data scarcity, humanoid robots are still a long way from being autonomous. Hardware is currently ahead of software – today’s robot body can handle much more dexterity than it did years ago (although it has reliability issues, as we saw with robots that collapsed in human marathons), but the brain is still nascent.”

Safety is a major hurdle for humanoid robots as well. A single high-profile incident could spark a public backlash, and China will likely consider how to quickly roll out this technology without moving too quickly. As the industry matures, more regulations are expected.

Given the lack of data, Zhao believes demand for humanoids will first grow in somewhat contained workplaces.

“Early momentum will likely be in industrial manufacturing, warehouse logistics, and retail, where tasks are repetitive, hours are long, and processes are straightforward — creating real demand and ideal conditions for humanoid robots to deliver value at scale,” he said.

Other players in the Asia-Pacific region

The development of humanoid robots is not a race between two countries. Japan’s robotics ecosystem — from startups to semiconductor heavyweights — is targeting mass production of robots by 2027. Japan, which has long been a leader with projects like Honda’s Asimo, Murata Manufacturing’s Murata Boy, and SoftBank Robotics’ Pepper, relies on precision and advanced control. One area unique to this nation is the increasing use of humanoid robots in elderly care.

James Rennie, CEO of Coral Capital, which invests in technology companies in Japan, believes Tokyo will continue to thrive in the humanoid robotics industry. “There are three factors likely to drive adoption of robots in Japan. The first is labor shortages and a desire to rely less on mass immigration. The second is the widespread cultural view of robots as our friends – more Doraemon vs. Terminator. The third is that Japan already dominates many parts of the robotics supply chain.”

Hyundai Motor’s Boston Dynamics unit has introduced a new Atlas Humanoid for factory use by 2028, with plans to produce up to 30,000 units annually in the United States as part of its push into artificial intelligence-driven robots.

However, for China, government policy, industrial strategy, labor shortages, and private capital are all converging to drive the country’s humanoid robotics momentum.

“Chinese leadership is best understood as a broad speed advantage,” Zhao said. “The ecosystem here compresses the entire cycle — R&D, supply chain, manufacturing, integration, and customer deployment — into a very tight loop. This means that human-powered companies can go from prototype to real-world deployment faster, learn from real operations, and iterate at a pace that is difficult to match elsewhere.”