Recently, the 2026 World Robot Conference concluded in Beijing. Robots invested in and developed by multiple mainstream automakers, including Chery, GAC, and Dongfeng, made a concentrated appearance. "Automakers crossing over to build robots" has become one of the most eye-catching highlights of this conference.
In the capital market, on August 24, XPeng's robotics business completed its first round of financing of over $900 million, with a post-investment valuation exceeding $6.3 billion, setting a new record for a single round of private equity financing in China's embodied AI industry.
From "four wheels" to "two legs," from building cars to "building humans"—this track is no longer just a sporadic trial by automakers. Relying on the advantages of technological homology and supply chain interoperability, the automotive industry and the humanoid robot industry are entering an unprecedented stage of integrated development.
Automakers Under Pressure: Robotics Track Becomes a Breakthrough
Against the backdrop of increasingly fierce competition in the NEV (New Energy Vehicle) industry and continuous pressure on vehicle profits, relying solely on car sales is no longer sufficient to support the long-term development of enterprises. Consequently, the humanoid robot industry is increasingly viewed by more companies as the next growth curve with immense potential.
It is understood that mainstream automakers both domestically and internationally, including Tesla, BYD, XPeng, Li Auto, GAC, Changan, Chery, SAIC, FAW, Dongfeng, and Geely, have all clearly laid out their strategies in the humanoid robot track.
Tesla is the pioneer of this wave. In 2021, Elon Musk first introduced the Optimus concept. Currently, the third-generation Optimus from Tesla has completed the debugging of its walking capabilities. The original Model S/X production line at the Fremont factory has been converted into a dedicated Optimus production line, with a planned annual capacity of 1 million units.
XPeng Group laid out its strategy early. Its established robotics research team has launched the humanoid robot IRON, which will enter the stage of scaled mass production by the end of this year. It will be officially launched in 2027, targeting external customers in retail and service industry scenarios, and will begin scaled deliveries in China and overseas. Next year, the monthly production capacity will rapidly increase to thousands of units based on market demand.
On March 27, SAIC Motor's first embodied AI humanoid robot employee, "Nengzai No. 1," officially took up its post on the mass production line for the Buick Zhijing E7 battery. This is SAIC Motor's first, and also one of the first in the Chinese automotive industry, cases of a humanoid robot truly being put into application on a mass production line.
GAC Group entered the humanoid robot track by incubating Huilun Technology. Its fourth-generation wheel-legged humanoid robot, GoMate Mini, has deployed nearly 50 units across 7 representative projects. Through a combined model of "sales + leasing," the company secured orders worth nearly 10 million RMB within half a year of its establishment, proving that its business model is viable.
At the Robot Conference, Mojia Robotics, under Chery, showcased its products, including the Smart Police Robot, Smart Care Robot, brain-controlled smart wheelchair, Moyin, and Mobo, covering multiple directions such as traffic management, medical services, public services, and family companionship.
NIO CEO William Li announced that Ren Shaoqing, NIO's Senior Vice President and head of the intelligent driving business, has founded an independent company focusing on physical AI foundation models and embodied AI. NIO will make a strategic investment in it and carry out cooperation. Leapmotor stated that it has already made plans in the field of humanoid robots and will announce relevant specific information in the near future.
Currently, mainstream domestic automakers have all completed their track layout and are accelerating their rush into the robotics track. Many state-owned automakers are even based on their own scenarios, taking the lead in landing factory pilots, enabling humanoid robots to undertake tasks such as material transport within the factory, equipment inspection, and auxiliary assembly, thus initiating real-scene applications.
Challenges Remain: Commercialization Still Awaits a Breakthrough
The core confidence for automakers flocking into the robotics sector comes from the high homology of underlying hardware technologies. The three major hardware systems—motors, motor controllers, and batteries—can be shared, and hardware costs can be continuously diluted through scaled mass production. Additionally, autonomous driving perception algorithms and large models can also be applied to robot motion control.
According to calculations by industry analysts, the overlap in hardware supply chains between the two exceeds 60%, and the AI software reuse rate for some automakers is as high as 70%. The mature vehicle manufacturing systems and mass production experience of automakers, along with offline channels such as 4S stores, have significantly lowered the threshold for the industrialization and landing of humanoid robots, representing a natural advantage that other technology companies do not possess.
Despite the broad industry prospects, for automakers, the current humanoid robot industry still faces obvious technical bottlenecks. Issues such as an immature embodied AI algorithm system, weak environmental generalization capabilities of robots, reliance on foreign sources for some core components, and the need for cost reduction in mass production remain the biggest obstacles to industrialization.
At the same time, humanoid robots are still some distance away from truly becoming a new profit growth point. Currently, most humanoid robots are still mainly applied in relatively closed scenarios such as industrial manufacturing, logistics and warehousing, and scientific research and education. In contrast, although home services are regarded as one of the largest application scenarios in the future, whether it is elderly care, housework, or child companionship, there is still a long way to go before creating commercial value.
Furthermore, from a comprehensive comparison, the essential differences between cars and robots are very obvious. Cars are standardized mobility hardware with relatively unified road environments, making it easy to form unified product standards and achieve rapid popularization. In contrast, the application scenarios for humanoid robots are extremely fragmented; the demands in industrial, commercial service, and special operation scenarios are completely different, making it difficult to achieve large-scale popularization relying on a single standardized model. The core competition for cars focuses on hardware range, chassis performance, and in-vehicle intelligence, whereas the core bottleneck for robots lies in the embodied AI software brain. Hardware is merely the carrier, and the difficulty of software algorithm iteration is far higher than that of in-vehicle intelligent systems. All these require further exploration by automakers. He Xiaopeng, Chairman and CEO of XPeng, once stated that the technical difficulty of advanced general-purpose humanoid robots is "at least 20 times higher" than that of smart cars.
From an industrial perspective, the entry of automakers brings a dual impact: for the automakers themselves, entering the robotics track is an extension and reuse of existing technologies and manufacturing capabilities, reducing R&D investment costs for new businesses and opening up a second growth curve. Self-developed robots can be deployed in the automakers' own vehicle factories, reducing long-term production labor costs and achieving "self-production for self-use and two-way empowerment."
For the robotics industry, automakers bring mature large-scale mass production quality control systems and massive upstream component supplier resources, which can accelerate the cost reduction and speed up the hardware of complete robot machines. At the same time, automakers come with mature globalized sales channels and overseas after-sales operation and maintenance systems, laying a channel foundation for domestic robots to go global. However, the entry of automakers also brings internal competition pressure within the industry. With larger capital volumes, automakers will squeeze the survival space of small and medium-sized native robot complete machine enterprises, forcing them to give up the complete machine track and turn to specialized, refined, distinctive, and innovative (SRDI) tracks such as segmented components and vertical scenario solutions, thereby promoting the refined division of labor and development in the industry.
Overall, the entry of automakers into humanoid robots is a long-term strategic bet anchored in the future. Although this path is full of challenges, it has already become the only way for Chinese automakers to break through growth bottlenecks, expand their second growth curve, and seize the AI era.
-END-
Editor/Layout: Zhu Wenfeng
Proofreader: Mei Yaxin
Supervisor: Liu Qicheng