In the bustling data factories of China, workers are tasked with teaching humanoid robots mundane, repetitive movements. These trainers, like Kim, a 20-year-old computer science major, don their virtual reality headsets and exoskeletons to mimic actions such as opening microwaves, folding clothes, or stacking wooden blocks. Their efforts are crucial to China's ambitious plan to become a robotics powerhouse, seen as the next battleground in the U.S.-China tech rivalry. The Chinese government has prioritized embodied intelligence, or artificial intelligence in physical form, as a key driver for economic growth, sparking a surge in investments for robot development, including humanoid robots capable of mimicking human behavior. The success of training large language models with massive data sets has led to the belief that a similar approach could work for robots, but robotics demands far more complex data, including visual information, joint motions, and rotations, which cannot be easily scraped from the internet or synthesized. To address this data crunch, local governments are establishing state-funded 'robot training centers,' where trainers like Kim act out everyday tasks repetitively to generate critical movement data. By December, over 40 state-owned robot data collection centers had been announced, with about two dozen already operational. One of the largest facilities, a training camp in suburban Beijing, covers an area of over 10,000 square meters and provides 16 specific scenarios for humanoid robot training, mimicking settings like car assembly lines and smart homes. However, concerns about overcapacity are growing, as the worldwide market for humanoid robots could reach $38 billion by 2035, with estimates of 250,000 shipments as early as 2030. Chinese and American companies are leading the race to develop intelligent robots that can live with humans and perform manual labor and household chores. The cost-effectiveness of China's data factories remains uncertain, as researchers explore alternative approaches to building fully intelligent robots, such as digital simulation or collecting data from existing robots. Despite the challenges, China's government support and strategic investments in robotics are expected to drive the industry forward, despite the risks of resource waste and overproduction.
