China Accounts for 97% of Global Humanoid Robot Shipments as Morgan Stanley Projects $5 Trillion Market
nashnova research
Chinese firms accounted for 97% of global humanoid robot shipments in the first half of this year, staking a claim on a market Morgan Stanley projects could exceed $5 trillion by 2050 — yet of roughly 20,000 units installed worldwide, only a few thousand are actually working; hand dexterity remains the core bottleneck.
Where does the 97% dominance come from?
According to Bloomberg, Chinese companies shipped 97% of all humanoid robots globally in H1 this year.
This means → virtually every humanoid robot reaching a customer came from a Chinese factory.
The foundation: over a decade of national strategy plus a complete hardware supply chain — motors, reducers, final assembly — all concentrated domestically.
Why do robot hands matter so much?
McKinsey partner Ani Kelkar: "Without dexterous hands, robots can only do limited tasks like pushing carts or moving large objects." Hand precision is the core barrier to commercialization.
In plain terms = a robot that walks but can't turn a screw or pick up a cup is a showpiece, not a worker.
Beijing startup LinkerBot (灵巧机器人) holds over 80% of global top-tier robot hand shipments, with clients including Samsung, Meta, and Siemens; it plans to double monthly capacity to 10,000 units.
Shipped ≠ actually working?
Roughly 20,000 humanoid robots have been installed worldwide, but only a few thousand are performing real tasks; the rest are largely for entertainment displays.
This means → the industry is in a "can build, can sell, but can't yet deploy at scale" phase — leading in shipments is not leading in commercialization.
This reflects a gap between hardware output and real-world utility: software, scenario adaptation, and hand precision all stand between shipping a robot and replacing a human worker.
The US–China split: hardware strength vs the "brain"?
China leads in hardware; the US is accelerating on AI models — the robot's "brain." Nvidia is investing heavily in physical AI; Musk's robotics project is advancing fast.
US capital markets far outpace China's fundraising capacity: SpaceX recently raised $75 billion. Chinese robotics leader Unitree (宇树科技) surged 460% on its August IPO debut but has since fallen roughly 50%.
In plain terms = China wins on "build fast, build cheap"; the US wins on "smarter brain, deeper pockets." The race hinges on which side closes its gap first.
Why is China's NDRC issuing a warning?
China's National Development and Reform Commission has warned that too many companies are blindly piling into the humanoid robot sector.
This means → regulators worry not about the industry itself but about herd-mentality investment creating overcapacity and wasted resources.
This reflects the tension between market enthusiasm and regulatory caution — whether China's robot industry can move from hardware shipments to genuine commercial deployment is the key test ahead.
市场有风险，内容仅供研究参考，不构成投资建议。