Chinese robotics shift from "technological breakthroughs" to "market explosion," valuation logic changes

The capital markets have shifted from chasing technology narratives to focusing more on real-world scenario needs and order conversion.

“China’s embodied intelligence industry is currently at a pivotal turning point, shifting from ‘technology breakthroughs’ to ‘market breakout.’”

During the 1st Overseas Investment and Comprehensive Services Fair (OIF 2026, hereinafter “Haiqie Fair”), Zhang Yimin, Chief Scientist at Intel China Research Institute and Chief Scientist at the Advanced Integrated Research Institute of Shanghai Jiao Tong University, said at the KOTON Robot Industry Ecosystem Forum held during the event. He said that although the industry as a whole is still facing many challenges such as technological maturity, ethics and privacy, as well as costs and market acceptance, China—leveraging advantages in the full industrial chain and market scale—is establishing a leading position in the global embodied intelligence race.

Industrial leapfrogging: from technology breakthroughs to global competition and coordination

At the ecosystem forum mentioned above, Liu Xiaoming, invited vice chairman of the China International Investment Promotion Association, said in a keynote speech that the robotics industry is entering a period of three major “headwinds” in policy, technology, and the market, and that China has the world’s most complete supply chain and the richest application scenarios. The key to future development lies in higher-level innovation platforms, more efficient overseas expansion channels, and tighter coordination within the industrial ecosystem.

Liu Xiaoming put forward three recommendations—for example, adhering to a policy-oriented approach and focusing on technological innovation; strengthening ecosystem coordination and growing stronger by working together; and maintaining a global pace and advancing high-quality internationalization. In the process of globalization, it is necessary to connect systems such as e-commerce, overseas channels, and cross-border services, so that China’s robots can go global with high quality, compliance, and strong branding, and build global discourse power.

In Zhang Yimin’s view, there is no need to wait for general artificial intelligence to become fully mature; solving 60%~70% of problems in specific tasks is enough to enable practical deployment of robots. At present, robot shipments are mainly for research purposes. In the future, they will expand into multiple scenarios such as industrial use, business services, and home elder care, and it is expected that robots entering the home elder care field may become a reality in the next 3~5 years.

Zhang Yimin said embodied intelligence has four major core directions that still need to be overcome: developing embodied intelligence large models and world models; optimizing real-time inference on the edge device; upgrading the robot’s body; and cloud-edge integration and human-robot collaboration. He emphasized that exporting embodied intelligence is absolutely not just about selling products; it must deeply integrate with local service systems and share after-sales resources to reduce enterprise costs.

Zhang Xin, a smart hardware analyst at Dolphin Investment Research, said that domestic companies, benefiting from shared technology origins, cost control, and speed advantages, occupy an important position in the upstream robot components sector, but there is a situation of “strong hardware and weak software.” Future opportunities will be concentrated in four directions: lowering costs through localization of hardware; breakthroughs in emerging hardware fields; tackling software large models; and “going overseas.” He suggested replicating the overseas expansion path of new-energy vehicles, but it is necessary to assess risks related to geopolitics, the supply chain, and human resources.

Kang Jingwei, chairman of the KOTON Group, said that the AI industry follows the “energy—chips—infrastructure—large models—applications” “five-layer cake” theory. As the top application of AI in the physical world, embodied intelligence robots must connect the entire industrial chain. China’s robot companies have been international from the day they were founded, and should leverage global technology and local industrial chain advantages to serve global markets.

Repricing: capital favors “hard-core” capabilities that can make money

Against the backdrop of the industry moving into a breakout period, the valuation logic in capital markets is undergoing a profound shift—namely, moving from chasing technology narratives to paying more attention to real scenario demand and order conversion.

In the investment and financing roundtable at the above-mentioned ecosystem forum, Wang Jiawei, founder of Ze Na Cheng Consulting, opened with data: According to incomplete statistics, from 2026 to today, the robotics segmentation sectors have completed more than 190 rounds of financing, with amounts exceeding 20 billion yuan. After setting a historical high in 2025, there is almost no suspense that 2026 will set another record as well.

Liu Yang, vice president of PIX Moving, proposed that the industry has shifted from “valuation using expected future revenue” to “pricing based on past stable revenue.” Capital requires that the financing funds must be used for industrialization and practical deployment, and for generating orders. He emphasized that in city service scenarios—with their high repetition and structured characteristics—they are becoming a high-quality entry point for industrializing embodied intelligence. This can accelerate technology deployment through “scenario dimensionality reduction.”

Chen Zhen, founder of Qian Chuang Capital, said that early investment hot spots focused on body or platform companies, but localization of core upstream components is the foundation of the industry; technological breakthroughs in areas such as reducers and sensors have more long-term value. Chen Zhilin, responsible person at 共熵 Investment, reminded that it is necessary to be wary of hardware valuation bubbles in high-end embodied intelligence robots. Industrial customers especially focus on cost recovery within 18 months; only companies that can access real demand scenario will have investment value.

Regarding the investment preferences of different capital players, it is generally believed in the industry that US-dollar and technology funds focus on frontier technologies such as tactile sensing sensors and simulation mechanics. Industrial capital focuses on replicable scenario deployment. Government guidance funds focus on industrial deployment and employment generation, and pay more attention to projects such as data factories. Looking ahead, the closer to customers a company is in the chain, the stronger its pricing power will be, and deep collaboration between upstream and downstream will become an inevitable trend. Investment hot spots will concentrate on scalable shipment scenarios, independent data providers, and industrial mergers and acquisitions. Compliance capabilities will become a company’s core competitive advantage.

Xu Renjie, partner at Jinghe Law Firm, said that before corporate financing, companies place more emphasis on building refined and detailed architecture. Cross-border architecture planning and the design of drag-along and guarantee (bet) terms are becoming even more rigorous, and founders’ joint liability is generally capped. To this end, it is recommended that companies build a three-dimensional intellectual property protection system, and proactively plan for the prevention and control of data compliance and AI ethics risks.

Massive information, precise analysis—right on the Sina Finance APP