Summary of Key Points
As a leading company in the field of humanoid robots, Yushu Technology went public on the STAR Market three months ahead of schedule. On its first day of trading, the stock price soared before quickly falling, reflecting the market's high expectations for its “universal robot platform” versus its short-term performance. Behind this listing is a clash between the urgency of capital to realize its valuation as soon as possible and the need for technology to be gradually verified according to engineering principles. Currently, the enthusiasm for the embodied intelligence industry far exceeds the maturity of the technology. Although robots have been mass-produced at an early stage, most are still used for scientific research and testing, and there is a long way to go before we see truly practical and commercially viable universal robots. Listing is not the end, but the beginning of a public validation phase. Ultimately, the success of the industry will be determined by the robots' ability to work stably and by the revenue generated from real applications.
I. What’s Behind the Dramatic Stock Price Fluctuations of Yushu’s Listing?
On the first day of trading, Yushu’s stock opened at 1,100 yuan (more than six times the issue price) before falling to 603 yuan. Such large fluctuations are not due to a sudden decline in the company’s performance. There are two main reasons:
1. Low number of tradable shares: The total number of shares is over 400 million, but only about 30 million are tradable (7.4%). It’s like there’s only a small basket of apples available, and naturally, the price goes up as everyone tries to buy them; it falls when people start selling.
2. Valuation is based on the future: Yushu’s current profits are not high (net profit of 288 million in 2025), but its market value is over 24 billion yuan, which means it would take more than 800 years to recoup the investment based on current profits. This investment is a bet on its potential to become a “universal robot platform” that can perform various tasks, accumulate data for AI model training, and even transform the labor market. Therefore, the stock price fluctuations reflect the market’s ongoing adjustment of its future expectations, and a few days of changes do not indicate the long-term prospects of the industry.
II. Yushu Can Produce Robots, but What’s Missing to Make Them “Universal”?
Yushu is stronger than many of its competitors in that it already has real products and revenue (5,500 humanoid robots were sold in 2025, generating 1.7 billion yuan in revenue), and it has reduced the cost of robots through mass production (from 260,000 yuan to 167,000 yuan). However, it still lacks a crucial element: intelligence capabilities. Its robots cannot handle various unfamiliar tasks as flexibly as humans, such as picking up different objects or adapting to complex environments. The company’s self-developed universal intelligence model is still in the pilot stage and has not been widely used. A common issue in the industry is that robots lack the ability to generalize (i.e., the models are not effective in different contexts), their hands are not nimble enough, they tend to break down easily, and their battery life is short. As a result, half of the funds raised from the listing will be invested in model development to address this weakness.
III. Is the Crazy Investment in Embodied Intelligence a Good or Bad Thing?
In the past year, there has been a surge in funding for embodied intelligence, with over 70 billion yuan invested in 2025 and another 50 billion yuan in the first half of 2026. While this capital inflow has its benefits, it also brings challenges:
Benefits: More money and resources have led to lower robot prices (e.g., Yushu’s robots), a more mature supply chain for core components (such as motors and sensors), and more robots being tested in factories and warehouses. The industry has moved from proving that robots can move to proving that they can work consistently.
Challenges: The rapid investment has outpaced the industry’s maturity. Many robots are still at the prototype stage, and few are actually profitable. Investment institutions are becoming more pragmatic, focusing on whether products are delivered, whether customers will repurchase them, and whether they can generate revenue in specific applications. After all, robotics is a long-term industry that requires time for technology and applications to mature.
IV. Many Robots Have Been Sold, but How Many Are Really Useful?
In 2025, 18,000 humanoid robots were sold globally (with China accounting for 95%), and another 22,000 were sold in the first half of 2026. However, most of these are not used for practical work:
- Over 85% are used for scientific research and education (school experiments), demonstration and interaction (exhibitions), and reception (malls).
- Only 13% are used in industrial manufacturing, and 5% in logistics and warehousing.
Why? Factories require robots to be reliable, accurate, and efficient—with a success rate of over 99%, no failures during continuous operation, and low maintenance costs. Current robots do not meet these criteria. For example, Yushu’s robots may fall several times during factory trials and cannot handle complex tasks. The industry is still in the transition from “technically feasible” to “commercially viable.”
V. Listing Is Not the End, but the Beginning of a Public Test
Yushu’s early listing has secured over 6 billion yuan for research and development, which is positive. However, it now faces public scrutiny:
- It previously communicated with investors privately, but now it must regularly release financial reports, making its performance and research progress transparent.
- Stock price fluctuations will force the company to balance short-term performance and long-term investment in research and development.
- Capital can speed up the acquisition of computing power and recruitment of talent, but engineering validation cannot be rushed. Robots need to go through countless trials, failures, and mistakes to become stable. It’s like a child learning to walk—only after falling enough can they learn to run.
In summary, listing is not the goal; it’s the beginning of a public evaluation process. The industry’s success will be determined by the robots’ ability to perform effectively and by customer willingness to continue investing. For the general public, short-term stock price fluctuations are less important than whether robots can truly enter factories and homes to solve real problems. Ultimately, only robots that can perform useful tasks are truly valuable.
Conclusion
Embodied intelligence is a promising but still growing industry. Capital wants quick results, but technology requires time to develop. Yushu’s listing is a microcosm of this: it has received funding but still needs to prove itself through time and practical application. For ordinary people, short-term stock price fluctuations should not be a major concern; what matters more is whether robots can become a practical solution in real-world scenarios. After all, only robots that can perform useful tasks are truly valuable.