Summary: China has launched a national ID system for humanoid robots, assigning each a 29-character code that tracks it from production to recycling. Over 28,000 robots across 200 models already have IDs. The system logs real-time performance data including joint wear, battery status, and AI training history.
China has launched a national identification system for humanoid robots. The Humanoid Full Lifecycle Management Service Platform, built by the Humanoid Robotics and Embodied Intelligence Standardization committee under the Ministry of Industry and Information Technology, assigns each robot a unique 29-character digital code that follows it from the factory floor to the scrapyard.
The code captures everything: manufacturer, product model, serial number, hardware specifications, AI capability level, software training history, and production records. It is modelled on China’s 18-character national citizen ID system but adds 11 extra characters to cover operational data specific to machines. More than 28,000 robots across 200 models have already been assigned an ID through the platform, which was launched by the Hubei province’s Humanoid Robotics Innovation Center in May.
This is not a static registry. The platform functions as a live digital record that tracks maintenance history, work environments, and real-time performance metrics including mechanical joint wear rates, battery degradation, and movement precision. When something goes wrong, the system is designed to enable rapid fault detection. When a robot is decommissioned, the ID follows it through recycling.
The scale of China’s humanoid industry explains why regulators moved now. The country has more than 100 humanoid robot manufacturers. Investment in robotics and embodied intelligence in 2025 exceeded the full-year 2024 total by the end of May, with China pouring $3.4 billion into new robotics ventures, 42 per cent more than the United States and five times Europe’s total. Shanghai has issued China’s first provincial plan for embodied intelligence, pairing R&D support with shared infrastructure for compute, testing, pilot production, and financing.
The robots are already showing up in the real world. A humanoid named Lightning completed the Beijing E-Town Half-Marathon in 50 minutes and 26 seconds earlier this year, beating the human world record by nearly seven minutes while navigating the 21-kilometre course autonomously. China’s State Grid Corporation plans to deploy 8,500 robots, including humanoids and robot dogs, for power grid operations. Tea farms in Hubei province have begun field trials with humanoid workers ahead of the 2026 World Robot Games.
The ID system addresses a governance gap that is widening as deployment accelerates. Without a standard way to track who built a robot, what software it runs, where it has been deployed, and how it has performed, liability becomes murky. If a humanoid injures a worker or damages property, regulators need a chain of information that connects the incident to a specific machine, its manufacturer, and its operational history. The ID code provides that chain.
There is a broader regulatory context. China has moved faster than any other country on AI governance frameworks, from algorithmic recommendation rules in 2022 to generative AI regulations in 2023 to deepfake and synthetic content rules in 2024. The robot ID system extends that approach to physical AI, treating humanoid robots as entities that require lifecycle oversight in the same way that vehicles, medical devices, and industrial equipment do.
The comparison to citizen IDs is deliberate but imperfect. Robots are not citizens. They do not have rights. The ID system is an industrial standard, not a legal status. But the structural parallel, a unique identifier issued by the state that tracks an entity across its entire existence, raises questions that other countries will eventually need to answer. As humanoid robots move from factories into hospitals, homes, and public spaces, who is responsible for what they do? The ID code does not answer that question, but it creates the informational infrastructure to start.
The United States and Europe have no equivalent system. The EU AI Act classifies AI systems by risk level but does not require individual identification of physical robots. The US has no federal framework for humanoid robot registration. China’s approach to AI governance has consistently prioritised state visibility into how the technology is deployed, and the robot ID system is the latest extension of that philosophy.
For manufacturers, the system creates both obligations and opportunities. Compliance means submitting detailed technical data for every unit produced. But it also means that a robot with a clean lifecycle record, well-maintained, regularly updated, deployed within its rated capabilities, carries a verifiable history that could function as a quality signal for buyers. In a market with 100-plus manufacturers and no dominant brand, standardisation is a competitive tool as much as a regulatory one.
The question for the rest of the world is not whether China’s approach is right but whether it is early. If humanoid robots become as common as China’s industrial policy intends, every country will need a way to track them. China is building that system while the rest of the world is still debating whether the robots are ready. The 28,000 units already in the database suggest the debate may be beside the point.
