AI Progress is a 'Critical Bottleneck' for Humanoid Robots, Says Unitree Founder
The founder of a leading Chinese robotics firm says AI technology has not yet reached a critical breakthrough, a global challenge he identifies as the main obstacle to the mass adoption of humanoid robots. He anticipates significant AI progress within 3-5 years but warns that widespread household integration will take much longer, with computing power limitations and talent shortages also constraining industrialization.
In an interview with the People's Daily published on August 13, 2025, Wang Xingxing, the founder of Unitree Robotics (杭州宇树科技有限公司), stated that while robot-related AI is advancing rapidly, it remains the core bottleneck for the industry. This insight comes as the humanoid robotics sector experiences a surge in interest, partly fueled by the high-profile public appearances of Unitree's own robots. Wang characterized the current excitement in 2025 as a "small spark," comparable to the early days of the internet, while underscoring China's manufacturing and cost advantages in the global race to develop the technology.
Below is the full English translation of the interview, preserving its original structure and content.
Maintain an Open Mindset Towards Fluctuations and Have Greater Confidence in the Future (Conversations with Entrepreneurs on 'The New') An Interview with Wang Xingxing, Founder of Unitree Robotics
Editor's Note: Since the beginning of 2025, Unitree Robotics' humanoid robots have frequently gone viral, and China's humanoid robot industry has received unprecedented attention. This has been met with praise, encouragement, criticism, and skepticism. Recently, at the Unitree Robotics headquarters in Hangzhou, Zhejiang, our reporter interviewed founder Wang Xingxing to hear the 1990s-born entrepreneur's response to public interest and his thoughts on the industry's development.
"In the long run, perhaps this is just a small spark, like the birth of the internet."
Q: Unitree Robotics' humanoid robots became incredibly popular after appearing on the CCTV Spring Festival Gala. Was this expected?
A: To be honest, it was completely unexpected. In 2021, our "robotic ox" was on the Spring Festival Gala once, and we originally estimated the effect this year would be similar to last time. We didn't expect the final result to far exceed our expectations; many people remembered the image of the robots dancing. In a sense, it became a landmark event combining traditional culture with high technology and also a cultural symbol. I feel very honored.
Q: What has been the impact on the company of this sudden surge in attention for humanoid robots?
A: The high level of attention is a good thing for the entire industry. Our company, related enterprises in the industry, including component suppliers, have all seen excellent performance in the first half of the year.
Of course, with the attention comes significant pressure. We have more people visiting, conducting research, and doing interviews. I personally oversee technology and products, so these activities have certainly taken up some of my time. For the industry to develop, it definitely needs attention. Of course, this process will have its ups and downs. Attention might be high today and wane tomorrow, but it will eventually return to rationality.
Q: Does this attention also come with dissenting voices?
A: It's normal for some people to like it and others not to. If a company or a product is liked by everyone, that would actually be abnormal.
Q: How far away are humanoid robots from being used in daily life?
A: We are still quite far from humanoid robots entering our daily lives and reaching every household. The industry is currently in its early stages. Therefore, we are starting with smaller applications, such as research and education, service and exhibition, simple industrial tasks, and as development platforms for AI companies.
Of course, our greatest hope and what we are currently working on is to have robots do work. They can perform various actions and help people with all kinds of tasks. However, the current level of technology is not sufficient for large-scale application in homes or factories. The fact that we are making gradual progress every year is a good thing, as it can foster greater understanding and tolerance for the industry.
Q: What is the biggest difficulty in achieving a large-scale rollout of humanoid robots?
A: The biggest challenge remains that the AI level of robots is insufficient; it has not broken through a critical point. This is a universal problem worldwide and something everyone is working hard on. The AI field can experience breakthroughs at any moment; technological leaps are quite common. A problem that seems unsolvable today might suddenly be resolved at some point in the future.
Q: How long do you think it will take to break through the bottleneck for scaled application?
A: The progress in robot AI technology is very rapid. For example, six months ago, if you had told me our robots could compete in fighting matches, I wouldn't have believed it myself. I would have thought it would take at least a year or two, but it was actually developed in just a few months. Of course, the industry's development is hard to predict. There may be significant progress within three to five years. However, I think entering ordinary households will be a relatively long process due to ethical and safety issues.
As long as AI technology continues to advance, the robotics industry will develop better and better. This is a chain effect brought by the AI industry, and the trend has been very clear over the past two to three years. People's imagination for AI is expanding, leading to a more optimistic outlook for the robotics industry. The humanoid robot boom in 2025, in the long run, might just be a small spark, like the birth of the internet in its early days.
Q: At this stage, what is the significance of having robots participate in various competitions?
A: As long as people enjoy it, providing emotional value is also a form of its value. This is also creating new demand. Robot performances and competitions will be an important part of the entire future robotics industry.
"We currently have relatively fewer AI talents, but in the long term, our competitiveness remains very strong."
Q: Unitree Robotics started with robotic dogs and has entered the humanoid robot race in the last couple of years. Why did you make this choice?
A: It is mainly due to AI technology. In the early years, AI was single-function, such as recognizing numbers, images, or faces. In recent years, the focus has shifted to multimodal, general-purpose AI that can do many things, which in turn requires a general-purpose robot platform, like the humanoid robots we are now making.
Q: Why humanoid robots?
A: The reason is simple: more people like humanoid robots than dislike them. This is the main reason, and the market is very rational.
Q: What is the global standing of China's humanoid robot industry?
A: We have manufacturing advantages and higher shipment volumes globally. Each country has its own strengths. At present, we have relatively fewer AI talents, but in the long term, our competitiveness is still very strong. Furthermore, because of our strengths in manufacturing and hardware, Chinese robot companies are more competitive on cost, leading to a lower barrier to entry. These factors are beneficial for future large-scale adoption.
Q: What are your thoughts on competitors and their products?
A: This industry is currently very hot, with many companies involved, each with its own advantages. It is important to maintain an open mindset. Healthy and reasonable competition is beneficial for the entire industry.
Q: How important is computing power to the humanoid robot industry?
A: It is very important, especially large-scale cluster computing power. A robot is a co-created entity with AI. However, robots, especially humanoid robots, run on batteries, and their battery load and power consumption limit their ability to carry large-scale computing power.
Q: How can the computing power problem be solved?
A: I think there will be a lot of distributed computing in the future. If a factory has a centralized computing center, the robots can connect to it. Such a centralized computing center is close by, so whether it's wired or wireless, the communication bandwidth is sufficient and the latency is very low. However, if a robot in Beijing connects to a computing center in Shanghai, the latency would be too high.
Q: Can the current talent supply meet the industry's development needs?
A: The humanoid robot industry has always been short of talent. This industry didn't exist before, and the relevant talent has only been gradually cultivated and developed in recent years. Additionally, the industry is still relatively niche, and the number of people who have truly entered it is not as large as one might think. It will take time for the quantity and quality of talent to match the industry's development needs.
Q: What types of talent are most in demand?
A: We lack people in all kinds of positions. Whether technical or non-technical, including business and HR, the most critical are AI talents.
Q: Have you considered collaborating with universities and research institutions?
A: We have always maintained close relationships with many universities at home and abroad. It is very important for us to promote technological progress in the robotics and AI industries and to build more connections with universities and young people.
Jointly advancing the industry's progress is the greatest value we have contributed to the entire industry over the past few years. In a sense, while a company's performance is one thing, it is more important to have a positive impact on the industry and society as a whole.
"Hoping more people, especially the young, will get involved and jointly advance the industry's progress."
Q: Unitree Robotics has previously published a book on quadruped robots and made some of its technology open-source. Why do this?
A: Actually, besides the book, we release some open-source projects every few months. We hope to promote the development of the industry. When the industry does well, the companies within it can thrive.
Another reason is to cultivate more talent. Whether it's through collaborating with universities, publishing books, or releasing open-source projects, we hope that more people, especially young people, will participate and jointly advance the industry's progress.
Q: Why is it difficult to rapidly cultivate R&D talent?
A: The industry is changing very quickly. The most important thing is to maintain the habit of learning. I do this myself. During the day, there are many company matters, but at night, I have a bit more personal time to read some papers and write some code, maintaining a state of learning and progress.
Q: How do you view the relationship between R&D and investment?
A: Whether learning a new technology or developing a new product, the most important thing is to get the direction right. Most R&D investment is spent on exploring the right direction, which is the most 'expensive' part.
It is also necessary to maintain a keen sense for technology and products. Being sufficiently perceptive will help you avoid many detours. It's important to use the lowest possible cost to explore, identify directions that are truly worth investing in, and then focus your learning and investment there.
Q: What is your ultimate vision for the development of the robotics industry?
A: To truly advance technological progress and take human productivity to a new level. The invention of the steam engine and electricity brought about immense changes to the productivity of human society. Currently, general-purpose AI and general-purpose robots could bring about an even greater change, allowing humans to gradually break free from strenuous physical labor.
Q: Unitree Robotics was founded in Hangzhou. What is the relationship between a city and a tech company?
A: Zhejiang has a very open and inclusive business environment. The development of a tech company is closely linked to the young people who have mastered emerging technologies. You could say that a city that is attractive to young people will certainly not lag in the development of its emerging industries.
Q: What other support does the robotics industry need?
A: I think the state is already very supportive. I hope everyone can embrace new technologies and products with a more open mind. Looking at the industry's development cycle, fluctuations are inevitable. Many people might be optimistic today, but things could look different tomorrow. Maintaining an open mindset towards these fluctuations and having greater confidence in the future is the greatest support for this industry.
Source: People's Daily, August 13, 2025 (Page 07)