Robot's ChatGPT moment approaching, says Unitree CEO

Wang's remarks draw a direct parallel to OpenAI's ChatGPT, which the body identifies as the catalyst for the global AI boom since its 2022 release and the branching point that drove mass adoption. The analogy is instructive: just as ChatGPT lowered the barrier to generative AI by making it accessible via natural conversation, Wang is predicting an equivalent inflection point for embodied AI—robots that can operate in novel real-world settings without extensive pre-programming. The body frames this as a "world model" capability: AI that lets robots understand and move through physical environments with human-like flexibility.

The 80% task-success target Wang cites is grounded in a practical scenario—a robot placed in a stranger's home following only voice or text instructions. This suggests the breakthrough he anticipates is not primarily about raw AI capability, but about the generalization gap: robots moving from controlled lab conditions or specialized deployments to arbitrary domestic spaces. The timeline he offers (2–3 years optimistic, 5–10 years realistic) reflects genuine uncertainty, but his framing at a major international robotics conference signals that industry leaders view this transition as inevitable rather than speculative.