Are we entering a humanoid robot winter? China’s warning may be the first domino of the upcoming
Aaron Plasser: Director of the ASTM International Robotics and Autonomous Systems Program
November 30, 2025
China rarely issues public market signals without reason, so the recent comments from the country regarding the humanoid robot industry are particularly noteworthy. At a press conference held in Beijing, the National Development and Reform Commission (NDRC) issued a clear warning: the development of the humanoid robot industry is too rapid, overly repetitive, and lacks depth. Currently, there are over 150 companies in China producing humanoid robots, many of which have designs that are astonishingly similar, with overlapping functionalities and nearly identical value propositions. The NDRC warned of the risk of market oversaturation — in this field, imitation far exceeds true innovation.
For an industry still in its early stages, this moment is significant. Over the past two years, humanoid robots have generated unprecedented attention through viral demonstration videos, billions of dollars in investment, and the vision of general-purpose embodied artificial intelligence on the horizon. However, hype often transcends physical laws, economic realities, and certifications. China’s warning suggests that the global humanoid robot industry may be approaching a familiar turning point, where excitement begins to breed negative consequences.
This aligns with my earlier prediction this year about the impending “humanoid robot winter” — during this period, high expectations conflict with the slow and methodical work required to develop safe, reliable, and commercially viable humanoid robots. A technological winter is not a collapse, but an adjustment. When the number of market participants exceeds the number of buyers, when R&D costs exceed the value of final delivery, and when beautiful demonstrations cannot compensate for the shortcomings of actual deployment, the winter will arrive. China’s acknowledgment of market saturation may be the first domino of this trend.
The formation of a bubble has deep structural reasons. Today, humanoid robot startups around the world face similar dilemmas: high capital consumption, investors’ urgent timelines, and fierce competition filled with homogenized platforms. Many companies are far from achieving a repeatable return on investment in warehouses or factories. More companies have yet to overcome the complex safety and insurance challenges required for actual deployment. Almost all companies struggle to bridge the widening gap between machine performance in controlled environments and actual customer needs in unpredictable environments.
But the bursting of the bubble does not mean the end of the story. In fact, it often marks the beginning of the real story. Those companies that can survive the “humanoid robot winter” will be those that can resist the temptation of flashy gimmicks and instead focus on the seemingly mundane fundamentals: compliance, reliability, interoperability, safety, and unit economics. They will prioritize real customer environments over carefully designed stages, measurable outcomes over viral internet fame, and engineering specifications over racing to capture market share.
Therefore, China’s warning is not a harbinger of collapse, but a sign of maturity. Every transformative technology goes through a period of ambitious yet unfulfilled potential. The technologies that ultimately survive are those that can adapt to change, recalibrate, and commit more firmly to creating real value.
If this is the beginning of the humanoid robot bubble, it may also be the start of something healthier: an industry forced to grow.
