Editor's Note: The embodied intelligent robot industry now stands at a critical intersection of technological iteration and large-scale deployment, accelerating its integration into various industries. The entire sector is rapidly transitioning from the "technology validation phase" to the "commercial popularization phase," with the outline of a new round of growth becoming increasingly clear. Meanwhile, more practical propositions are placed before every industry practitioner: Where should technological breakthroughs be made? Where should the market be deeply cultivated? How should the ecosystem be co-built? In this context, on the occasion of the 2026 World Robot Conference, China Electronics News has invited representatives from multiple robotics companies to share frontline industry voices around core topics such as technological evolution, market deployment, ecosystem synergy, and future trends, leaving a vivid footnote for the next chapter of the robotics industry.
How do you evaluate the current development stage of the embodied intelligent robot industry?
Where do the greatest development opportunities come from?
Xiao Xiangjiang, Chairman of Chaoneng Robot
Embodied AI is currently in a leapfrog period from "technology validation" to "large-scale commercialization," with 2026 marking a watershed for industrialization. However, the general-purpose humanoid track is still in the early market. The true meaning of the watershed is crossing the "chasm": creating standardized products and winning batch repurchases from pragmatic customers. Opportunities arise from three aspects: in terms of policy, embodied AI is included in the six major future industries of the "15th Five-Year Plan"; technologically, the explosion of large models and world models ushers in the enlightenment moment for robots to transition from "understanding the digital world" to "acting in the physical world"; in the market, China's massive application scenarios are the most precious testing ground and data mine. This year, the industry presents four major trends: from showing off skills to practical implementation; from single-agent intelligence to cluster collaboration, where the robot "brain" will become the next commanding height; returning from general narratives to specialized deep cultivation, with generalized humanoid hardware and necessarily specialized software and non-humanoid forms; entering the scenario era from the product era, where whoever masters the scenario and channel entrances will master the industry discourse power.
Gao Jiyang, Founder and CEO of Xinghaitu
I believe we are in a critical turning point for a comprehensive leap from "technology validation" to "productivity realization". For a long time in the past, the industry's focus was on hardware iteration, prototype demonstrations, and single-point technological research, essentially answering the question of "whether it can move." However, starting in 2026, the industry has officially entered the core inaugural year of commercial deployment and large-scale delivery. The evaluation criteria have shifted from "whether it can be made" to "whether it can generate benefits"—with scenario deployment, efficiency realization, and commercial closed-loop becoming the main theme, and technological value beginning to truly transform into industrial productivity.
The greatest opportunity in this transition comes from the fundamental upgrade of the intelligence paradigm. As embodied large model technology matures and lands, the industry consensus has been thoroughly reshaped: hardware is the carrier, and intelligence is the soul. Only by enabling the continuous evolution of the "brain" can hardware release its true industrial value. Therefore, the market focus is shifting from "stacking hardware and competing on parameters" to "competing on models, algorithms, and scenario generalization capabilities," and companies focused on building the "embodied brain" are becoming the core driving force propelling the industry forward.
Since 2026, the industry has shown two distinct characteristics: first, leading enterprises are accelerating their transition from single-scenario pilots to multi-scenario large-scale replication, with sharply increased requirements for generalization capabilities and data closed-loops; second, capital and resources are clearly leaning towards the algorithm and model layers, making the trend of "software defining hardware" irreversible. This means that the key to future competition lies not in who owns more expensive joint motors, but in whose intelligent system can continuously learn, quickly adapt, and deliver stably in real-world environments.
Since its inception, Xinghaitu has anchored this ultimate trend, adhering to the core strategy of "one brain, multiple forms" without deviation. We have built a dual-technology route of VLA models and WAM world models. The two deeply collaborate to form a complete intelligence leap from "perception-decision-execution" to "prediction-optimization-closed-loop," building an industry-leading general embodied AI capability foundation. It is precisely this "brain-driven" system that enables us to steadily approach customers and realize value in 2026, the inaugural year of commercialization—which is also our most pragmatic response to the industry cycle.
Dr. Xu Chunshan, Founder and Chairman of Dexterous Robotics
Currently, the embodied intelligent robot industry is moving at full speed from the "proof of concept" and "prototype demonstration" stages to a critical turning point of "scenario deployment" and "large-scale commercialization." The industry is no longer competing in single demonstration skills, but in the ability of robots to work stably and continuously under real working conditions. The 2026 World Artificial Intelligence Conference clearly demonstrated this: the industry is moving from technological exhibition to application validation. A large number of enterprises have set up real-scenario applications, with real tasks such as industrial production lines, power operation and maintenance, and pharmacy sorting becoming the mainstream on the exhibition booths.
The greatest development opportunity comes from the resonance between the national top-level strategy and the real needs of the industry. Embodied AI has been elevated to a national strategic track. China possesses the richest manufacturing scenarios, the most complete supply chain system, and the most proactive policy environment in the world, providing a unique testing ground and iterative soil for embodied AI to evolve from being usable to being stably and reliably usable.
Zhang Dapeng, Assistant Vice President of Leju Robotics
Currently, the embodied intelligent robot industry is in a critical turning point of leaping from "technology validation" to "large-scale commercialization." Driven by breakthroughs in large model-powered embodied AI technology, the industry has stepped out of the laboratory and entered real industrial scenarios for testing.
The greatest development opportunities come from three aspects: first, the deep integration of AI large models and robot bodies, endowing robots with stronger generalization and autonomous decision-making capabilities; second, China's complete manufacturing system and rich application scenarios, providing natural soil for the deployment of embodied AI; third, continuous policy support, with the national level listing humanoid robots as a key development direction.
Since 2026, the industry has shown three significant trends: first, industrial manufacturing scenarios have taken the lead in breaking through, with humanoid robots beginning to enter production lines in automobiles, logistics, 3C electronics, and other fields to perform real tasks; second, full-stack in-house R&D capabilities have become a competitive watershed, highlighting the advantages of enterprises with independent and controllable full-link capabilities from "brain-cerebellum-body-data"; third, the industry ecosystem is moving from closed to open, with infrastructure represented by open-source datasets and developer communities accelerating the efficiency of scenario deployment.
As one of the earliest domestic enterprises to achieve mass production and delivery of humanoid robots, Leju has always adhered to scenario-driven technological iteration and has currently formed four major deployment scenarios: industrial, commercial services, scientific research, and training grounds. We believe that industrialization is not a sprint that happens overnight, but a long-distance run that requires patience and perseverance.
Yang Zhengye, Marketing Director of the National and Local Joint Humanoid Robot Innovation Center
Overall, current embodied intelligent robots can basically run as prototypes, but once placed in real production lines, their stability and adaptive capabilities still fall short, and there are no truly commercially mature solutions yet. The greatest opportunities come, on the one hand, from manufacturing upgrades, where factories face labor shortages and increasingly strong demands for flexible production; and on the other hand, from the integration of large models and robotics technologies, coupled with our complete domestic supply chain, massive industrial scenarios, continuous policy support at all levels, and the ability to draw on mature experiences from the smart vehicle industry, creating an excellent window of opportunity for the industry.
Lin Jiawei, CMO of Kuawei Intelligence
The current industry is in a critical turning point from technology validation to large-scale commercialization. 2026 is widely regarded by the industry as the "inaugural year of embodied AI applications." In the first half of the year, the total financing in the domestic embodied AI track has exceeded 90 billion RMB. From the government and capital to the industry, everyone is concerned about how to quickly commercialize and deploy this cutting-edge technology. We believe the greatest development opportunity lies in the fact that, from general intelligent manufacturing to commercial retail, cultural tourism services, and other fields, a large number of unstructured tasks remain unsolved by automation. These require more flexible, general, and generalized product technologies, which is precisely the most valuable main battlefield for embodied AI.
Deng Xiong, Vice President of Suzhou Veichi Electrical Co., Ltd.
Embodied AI is a core breakthrough for the next generation of artificial intelligence, driving the intelligent equipment industry to leap and upgrade from automation and intelligence to autonomy. Currently, the domestic industry is in a rapid development period driven by the dual engines of basic research and industrial applications: technologically, core algorithms continue to break through, and the perception, understanding, and execution capabilities of agents steadily improve; industrially, hardware forms are becoming increasingly diverse, and simulation and scenario deployment are continuously deepening.
The development opportunities for embodied AI are rooted in real industrial demands and technological iterations. China's manufacturing industry is currently leaping from "automation" to "autonomy." Traditional automated equipment is gradually struggling to adapt to the operational requirements of multi-variety flexible production and complex unstructured scenarios. The integrated perception, decision-making, and execution capabilities brought by embodied AI can precisely fill this gap. China possesses extremely rich manufacturing and service industry scenarios globally. Deeply cultivating refined vertical fields to provide diversified scenario solutions, continuously improving technological strength and product competitiveness, enhancing market responsiveness, and promoting the upgrading of the intelligent manufacturing industry—all of these represent excellent development paths for the vast number of industry practitioners.
Hu Zheqi, Think Tank Manager of Yuequan Bionics
Currently, the embodied intelligent robot industry is in a critical stage of "transitioning from the early stage of industrialization to partial large-scale application." The industry's focus has shifted from "whether a robot can perform a specific action" to "whether it can complete tasks in real scenarios over the long term, stably, safely, and at low cost." At present, structured and semi-structured scenarios such as manufacturing, logistics, inspection, and special operations have entered the stage of real-scenario training and normalized validation. However, the cross-scenario generalization capabilities, reliability, and economics of general-purpose products still need further improvement. Therefore, the current stage is closer to the eve of large-scale commercialization, rather than having entered a fully mature period.
The greatest future opportunities mainly come from three aspects: first, the real demands generated by the flexibility of manufacturing, demographic changes, and the replacement of high-risk positions; second, the accelerated integration of large models, multimodal perception, motion control, dexterous execution, and new materials, enabling robots to evolve from automated equipment to autonomous operating systems; third, China's complete manufacturing system, rich application scenarios, and rapid iteration capabilities, which have the conditions to form an industrial flywheel of "scenario opening—data accumulation—product iteration—large-scale deployment." For Yuequan Bionics, key links such as dexterous hands and tactile perception are not only the execution foundation for robots to complete complex tasks but also important entry points connecting digital intelligence with the physical world.
Yang Yuxin, Chief Marketing Officer of Black Sesame Technologies
Overall, the current embodied intelligent robot industry is in a critical transitional stage from technology validation to breakthroughs in large-scale commercial pilot applications. The core technology framework has gradually become clear, and the integration path of large models and robots is basically defined. Pilot projects in various scenarios are rapidly increasing, but overall, it is still dominated by small-scale deployment in segmented scenarios, and there is still a certain development cycle before large-scale generalized commercialization.
The greatest development opportunities come from three aspects:
First, the endogenous drive of AI technology. The rapid iteration of multimodal large models and edge-side inference technologies provides the technological foundation for upgrading robots from "automated equipment executing specific instructions" to "intelligent terminals with autonomous perception and decision-making capabilities." The improvement of edge-side computing power also provides core support for the deployment of large models in robots.
Second, the real pull of industrial demands. There is an urgent need for automation and intelligence upgrades in fields such as industrial manufacturing, smart logistics, and public services. Demographic changes are also driving the continuous improvement of the acceptance of embodied AI across industries, and real commercialization demands are gradually being released.
Third, the accelerated synergy of the industrial ecosystem. From chips and core components to large models and complete machine manufacturing, and then to scenario applications, the linkage between various links in the industrial chain is becoming increasingly close. The efficiency of upstream and downstream collaborative innovation continues to improve, providing systematic support for industrial breakthroughs.
Black Sesame Technologies has precisely identified this industry trend. Relying on its technological accumulation and forward-looking layout in automotive-grade AI chips, it has launched the SesameX Multi-dimensional Embodied AI Computing Platform, using mature edge-side computing power solutions to help the industry accelerate the transition from technology validation to commercial deployment.
Author | Editorial Department of China Electronics News Editor | Wu Lilin Art Editor | Ma Liya Supervisor | Lian Xiaodong