Unitree CEO Wang Xingxing on Robotics, Industrial Challenges, and Growth at 2025 Summer Davos
Over the past two days, the 2025 Summer Davos Forum has been taking place in Tianjin. Wang Xingxing (王兴兴), CEO of Unitree Robotics, participated in a panel discussion and shared his views on the future of commercial robotics, key industry challenges, and the current state of his company.
Wang Xingxing emphasized that the biggest challenge facing the robotics industry is safety, especially as robots begin interacting with humans in close-range, real-world scenarios like homes. While humanoid robots are not yet ready for mass deployment, he believes they will first be adopted in industrial and commercial settings and could begin performing tasks like room-cleaning or item delivery within 1–2 years.
He stressed that continuous innovation—particularly in motion control, mechanical design, and sensor integration—is essential to staying ahead in a competitive field. Reflecting on Unitree’s rapid growth, Wang credited both a booming industry and a supportive business environment in Zhejiang.
He highlighted China’s young generation as a key driver of AI progress and encouraged ongoing learning and experimentation. Wang also expressed optimism that AI model improvements will soon make robots significantly more versatile, and he welcomed collaboration with large partners like Haier—while cautioning that true market readiness depends on precise alignment between product capabilities and real-world needs.
(photo from SCMP)
Below are selected excerpts from his remarks:
Moderator: What do you see as the biggest challenge for the robotics industry right now?
Wang Xingxing: Our goal is to enable robots to truly get work done—especially the hard, repetitive tasks people don’t want to do. The household is a huge application scenario, but it’s also incredibly challenging, especially in terms of safety. We have to address those issues step by step.
Not long ago, one of our clients used a robot developed on our platform, and during an event it accidentally stepped on a little girl’s shoe. It became a hot topic online. Luckily, it was just the shoe and not her foot, and the robot didn’t fall or knock her down. But even that is a major red flag.
For products that interact with people at close range, safety issues may be even more critical than technical ones. Ethical and moral concerns also become more challenging.
Right now, it’s not realistic for robots to handle household tasks directly. However, we’re training them through activities like dancing and sparring to learn full-body movements, with the long-term goal of freeing human labor.
Although our product development isn’t limited to humanoid robots, AI training is currently based mainly on human data, so humanoid forms are easier to train and apply. In the future—once AGI truly arrives—robot forms will likely diversify.
In the past year, we’ve focused on enhancing our robots’ arm manipulation capabilities, so they can perform more complex tasks like part assembly or opening doors. This relies on coordinated innovation in mechanical design, motion control algorithms, and sensor integration.
Humanoid robots will likely first find applications in industrial and commercial scenarios, which are core to the next wave of the "industrial revolution." We’ll also need to think ahead about human-robot ethics and reskilling the workforce. Within the next 1–2 years, I believe robots will be able to do things like tidy rooms or deliver items in both commercial and home settings.
Moderator: What’s your company’s current annual revenue?
Wang: Around one billion RMB.
Moderator: How many employees do you have now?
Wang: It was just me when I started in 2016. Now, including manufacturing, we have over 1,000 people.
Moderator: Do you still get any sleep?
Wang: It’s not like I can’t sleep, but there’s a lot to do. The main thing is continuing to improve our product and technology. We weren’t founded recently—we started in 2016. Back then, the industry went through a wave of hype over robotic dogs and other forms. Today, the competition is more intense, but we have a much stronger foundation now—better products, broader capabilities. So I think we can handle the pressure.
Moderator: Sounds like you’re still very confident in your technology？
Wang: Definitely. The key is to keep improving. For companies like ours, if you stagnate—whether in product or technology—within a year or two, you’ll become mediocre. That’s inevitable.
In a frontier tech industry, you have to make visible technical progress every single month to stay ahead. If you stall, no one can really help you. That’s the reality. A lot of the pressure on me and the company is about keeping up innovation.
In recent years, China has created a very open environment for tech development. One of the biggest factors, in my view, is that the younger generation—especially those born in the '90s and 2000s—grew up with access to cutting-edge global technologies via the internet. They’ve had solid exposure and support, and society is investing more in young people. New AI technologies often emerge within five-year cycles, and what young people are learning in school now is the latest. So in a way, they have more opportunities and a broader stage to show what they can do. Given how fast things move in AI, they have a higher chance of building something truly new.
My advice: don’t panic. Keep learning, keep thinking. I myself have been studying AI lately. Robotics is one of the most anticipated areas in the AI community. I hope fundamental advances in AI models will make machines more versatile—whether in motion or reading and comprehension. Hopefully by next year we’ll see AI-powered humanoid robots entering factories.
Moderator: Unitree has drawn a lot of attention in recent years. What’s enabled this rapid growth since your founding in 2016? Has the city or local environment played a role?
Wang: For any company, the most important factor in doing well is that the industry itself is doing well. That is, there’s high demand, strong interest, and solid support. If you’re in a fast-growing field like robotics or AI, and you’re near the top of that field, you’ll naturally benefit. Also, the local environment—like Zhejiang province’s pro-business policies and deep talent pool—has helped a lot.
Another factor is that we’ve grown stronger operationally. We have more product lines, better performance, more talent. I’ve been in this field for years—started doing robotics around 2009. And globally, robots have now become one of the hottest areas, second only to AI.
While the industry hasn’t hit mass adoption yet, I’m optimistic that progress in AI models will drive broader market expansion. In fact, the pressure to keep innovating technologically is greater than the challenge of expanding into markets.
We’ve been actively advancing commercial applications and partnerships for humanoid robots, but the timing still isn’t quite right. AI and robotics develop in stages, and often plateau for a while before leaping forward. Our own experience shows that products only start selling better after breakthroughs in technology enhance their capabilities, which then unlocks new markets.
If you push into a market too early—before the tech is ready—you’ll struggle. In some cases, it could lead to heavy losses.
During the forum, Wang also shared his outlook for the next year:
“With improvements in AI models, I expect that within the next 6 to 12 months, humanoid robots will be able to play football better than humans. And by next year, we’ll see them in factories—and possibly even entering households on a large scale.”
During a panel discussion, Wang Xingxing was joined by Haier Group Chairman and CEO Zhou Yunjie(周云杰). Zhou spoke about Haier’s smart home ecosystem and vision for function-specific service robots tailored to real-life scenarios like elder care and rehabilitation.
Zhou emphasized that future competition will not be company-to-company, but ecosystem-to-ecosystem. Haier aims to integrate upstream and downstream partners to build broader industrial ecosystems in home appliances, healthcare, and industrial internet.
In response, Wang said he looked forward to working with Haier:
Unitree has always been working on partnerships with large groups like Haier,
but acknowledged that mass deployment of humanoid robots will take time and emphasized the need for technological readiness to align with market needs.