Unitree Founder Wang Xingxing: The Core Bottleneck for Robots Is Whether They Can Resolve Millimetre-Level Work Errors
Summary
Speaking at the 5th Global Digital Trade Expo in Hangzhou, Unitree founder Wang Xingxing said embodied AI will inevitably have its own "ChatGPT moment," with the tipping point being when robots can complete 80% of tasks via voice-driven embodied capability in 80% of unfamiliar scenarios. He said the biggest current problem is the mismatch between AI model inputs/outputs and the real physical world, leaving millimetre-level errors in robot work, and reiterated that humanoid robots' efficiency and capability remain insufficient for large-scale deployment in factories and homes.
Bubble analysis
This item is about robotics technical maturity, not about where AI investment money comes from or what returns it earns, so it bears only indirectly on the AI-bubble question. Still, Wang's admission that humanoid robots are currently less efficient than humans and cannot be deployed at scale implies that embodied-AI revenue has not yet arrived while investment keeps flowing in — the classic early-stage pattern of spending running ahead of revenue.
Related signals
linked by shared tags
- Behind the 4 Trillion Compute Network: What Does the Tech Foundation of 'New Quality Productive Forces' Look Like?#china
- Qineng Technology's Datong Super Energy Complex Tianzhen Computing Park Phase I Tops Out#china
- An education company selling compute? Doushen Education lands an RMB 800 million cloud computing services order#china
- Leading Optical Interconnect Player for Compute: Xihe Technology Discloses Second Round of Regulatory Inquiry#china
- Micron says robots will keep RAM prices high, even if the AI bubble pops#robotics