East Asian Technology Intelligence
Japan & China technology, translated and contextualized for Western readers
Industrial Automation & Robotics
星动纪元 / 北京星动纪元科技有限公司
A Beijing-based embodied-AI robotics startup developing humanoid robot bodies, wheeled service robots, and dexterous hands for research and industrial automation.
Robot Era is a well-funded Chinese embodied-AI company with a credible research origin and a broad humanoid product range. Its central tension is that while it presents evidence of hardware delivery and customer adoption, it does not disclose the financial or unit-economics data needed to establish whether those deployments are scalable and profitable.
Robot Era was founded in 2023 as a commercialization effort emerging from Tsinghua University's Institute for Interdisciplinary Information Sciences. The Beijing-based startup develops the physical-agent stack required for embodied artificial intelligence, including humanoid robot bodies, wheeled service robots, dexterous hands, and motion-control software. Its core product lines include the L7 full-size bipedal humanoid, the Q5 wheeled service humanoid, and the XHAND series of direct-drive dexterous hands.
Rather than competing in foundation models or semiconductor design, Robot Era provides the physical infrastructure that allows AI models to interact with the real world. The company serves research institutions, technology companies, and industrial users, positioning its hardware as a platform for both executing tasks and collecting real-world manipulation data. While it has secured substantial venture and state-linked financing, it remains a private, early-stage enterprise navigating a crowded domestic robotics market.
Robot Era earns revenue by supplying the physical embodiments and manipulation hardware required for AI models to act in the physical world. Its direct-drive dexterous hands, such as the XHAND series, are particularly important because they provide both manipulation capability and a critical source of real-world interaction data for robot-learning systems.
AI demand reaches the company through the physical-agent layer rather than through semiconductor sales or cloud computing. Research laboratories purchase its robot platforms and dexterous hands to train and evaluate embodied models. Industrial users purchase its robot systems for material handling, sorting, scanning, and related logistics tasks. The company also uses deployed systems and teleoperation to generate data that can improve manipulation and embodied-model performance.
Robot Era states that nine of the world's ten most valuable technology companies are customers, though it does not name all nine or disclose contract values. Public disclosures do not quantify revenue attributable to AI-model training, logistics, service robotics, or overseas customers separately, making its exact financial exposure to AI infrastructure difficult to measure.
Robot Era's technical proposition is an integrated physical-agent stack rather than a single humanoid chassis. It combines bipedal locomotion, wheeled service mobility, direct-drive dexterous hands, teleoperation, control software, and embodied-model work. Its direct-drive hand architecture reduces the mechanical transmission complexity of some conventional robotic hands and is presented by the company as uniquely suited to reinforcement-learning-based manipulation.
The company's practical differentiation is strongest where a task needs both whole-body movement and hand-level manipulation, such as industrial sorting or retail service. A competitor must replicate reliable hardware integration, safety performance, actuator and hand design, teleoperation workflows, data capture, and field validation.
However, these barriers are not unique to Robot Era. China's humanoid sector includes several well-funded companies pursuing similar hardware-software integration, and many customers can run pilots across competing platforms. The lack of disclosed patent totals, production yields, or unit economics limits external verification of its technical moat relative to better-disclosed rivals.
Robot Era depends on externally sourced components common to humanoid production, including actuators, motors, reducers, sensors, compute modules, batteries, high-performance semiconductors, and machine-vision components. Specific production suppliers for these inputs are not disclosed, though the company claims that more than 95% of its hardware is self-developed.
On the customer side, Robot Era sells or deploys products with research institutions, technology companies, and logistics users. Publicly named users or deployment counterparts include MIT, Stanford University, UC Berkeley, ByteDance Robotics Lab, Beizi Technology, Haier Smart Home, Lenovo, Century Golden Resources, SF Group, and China Post. The company also claims adoption by NVIDIA, Apple, and Boston Dynamics. Customer names are partly public, but Robot Era does not disclose a complete customer list, revenue concentration, or recurring-service revenue, leaving its downstream dependencies opaque.
Robot Era's robots require high-performance compute, GPUs, perception hardware, and potentially foreign-origin components that can be affected by United States export controls on China-bound advanced AI hardware. With more than 50% of its orders reported as coming from overseas customers in 2025, the company faces significant exposure to export licensing, customer procurement restrictions, data-security scrutiny, and cross-border service logistics.
Domestically, Robot Era benefits from its Tsinghua University origins and state-linked capital backing, which improve access to engineering talent, pilot sites, and enterprise budgets. However, this state-affiliated equity and industrial backing can increase geopolitical scrutiny for some overseas customers and partners. The company must balance its reliance on international research and technology customers with the realities of operating as a state-influenced Chinese robotics developer in a restricted technology environment.
| Risk | Severity | Why it matters |
|---|---|---|
| Advanced-compute export controls | High | Robots require high-performance compute and GPUs that can be affected by U.S. export controls. |
| Semiconductor supply dependence | High | Undisclosed chip and sensor suppliers leave exposure to procurement disruption difficult to quantify. |
| Private-company financial opacity | High | Lack of disclosed revenue and unit economics limits assessment of sustainable growth. |
| Humanoid-unit economics | High | Deployments remain largely pilots, and robots require expensive actuators, sensors, and maintenance. |
| Competition and price compression | High | Multiple Chinese humanoid developers compete for the same capital, components, and enterprise budgets. |
| Overseas-customer execution | Medium | High overseas order share creates exposure to export licensing, data-security scrutiny, and service logistics. |
| Customer concentration | Medium | Customer identities and sales shares are not fully disclosed, so concentration could be material. |
| Safety, liability, and regulation | Medium | Industrial humanoids operate near people, so safety incidents or regulations can delay deployments. |
Robot Era is a privately held company and does not publish audited annual reports, a complete cap table, board roster, or formal governance report. It is described as a Tsinghua University-held or Tsinghua-origin enterprise, with founder and CEO Chen Jianyu serving as an assistant professor at the university.
While not publicly described as a state-owned enterprise, it is state-influenced, with equity held by Tsinghua and participation from state-linked capital such as China Chengtong Fund. The reported investor base also includes CDH VGC, Haier Capital, Sequoia China, IDG Capital, and Geely Capital. Ownership percentages, board seats, and governance rights are not disclosed, resulting in high financial and operational opacity.
We haven't written about Robot Era yet. Subscribe below to get it when we do.
Compiled with AI-assisted research from company filings, market data, and published reporting as of September 14, 2026, then reviewed by AsiaAI.FYI. Figures marked Estimate are not company-reported. Check primary filings before relying on any number.
Confidence: D. Product lines, founder identity, financing rounds, and selected partners have public corroboration, but financial performance, unit economics, and manufacturing capacity are opaque because the company is private.
Main sources: Company website and product pages; Company and investor announcements; Chinese financial press; Technology-industry press; Research and robotics-industry publications; Private-company database entries.
East Asian Technology Intelligence
Japan & China tech news — translated, contextualized, and delivered for Western readers.
Free. Unsubscribe anytime.