OpenAI’s GPT-6 may lead to robotics breakthrough as soon as next year: Unitree founder

The founder of Unitree Robotics, a Chinese company, has revised his timeline for a significant advancement in embodied intelligence. Prior to GPT-6's release, Wang Xingxing had estimated the industry could take two to three years at its fastest, or as long as a decade, to reach a "ChatGPT moment" for embodied intelligence. The new model has prompted him to shorten this expectation to "within one or two years."

This revised forecast suggests that the rapid development of large language models like GPT-6 is seen by some industry leaders as a potential catalyst for robotics. The implication is that advancements in AI's ability to understand and generate complex instructions could translate directly into improved capabilities for physical robots, rather than requiring separate, slower progress in hardware or specialized robotics AI.

Wang's specific mention of a potential GPT-7 model next year "cracking the problems of embodied intelligence" indicates a belief that the next iteration of these AI models could provide the necessary computational or reasoning frameworks to overcome current hurdles in robot interaction with the physical world. This perspective emphasizes the software-driven nature of the anticipated breakthrough, where general AI models become key enablers for robotics.