ACE Robotics: Embodied AI Expected to Reach 'ChatGPT Moment' by Late 2027, Data Bottleneck Identified as Key Variable

BlockBeats News, August 23: During an interview with Reuters, Wang Xiaogang, Chairman of ACE Robotics, a China-based embodied AI startup, stated, "We expect that by the end of next year, driven by the world model and environmental data capture, embodied AI will embrace its ChatGPT moment." ACE Robotics, backed by Ant Group and SenseTime, raised over $100 million in the first half of 2026 and plans to launch an IPO "as soon as regulatory conditions allow."

Wang Xiaogang believes that current robots still struggle to reliably perform a wide range of tasks in unfamiliar environments. The key constraint is not the algorithm, but the severe lack of training data. "The entire industry has accumulated only about 100,000 hours of data over the past few years, far from enough to train an embodied base model." ACE Robotics plans to collect tens of millions of hours of real-world environmental data within two years and deploy the technology in 1,000 stores in the next year.

In the robotics industry, Boston Dynamics released the production version of Atlas in January, claiming that AI has brought it closer to commercial deployment. In June, Alibaba launched the Qwen-Robot Suite robot model kit, and researchers are experimenting with data collection solutions like the HumanoidExo exoskeleton. The competition in embodied AI is shifting from hardware prototypes to a dual showdown of data scale and model capability.