Unitree chief says humanoid AI's 'ChatGPT moment' arriving in 1-2 yrs

Artificial intelligence (AI) is developing faster and faster around the world. This has started a new debate about whether human-like robots are getting close to their "ChatGPT moment," according to the head of Chinese robot maker Unitree Robotics.

Wang Xingxing, the founder and chairman of Unitree Robotics, which is based in Hangzhou, Zhejiang province, now believes this big turning point for AI-powered robots could happen sooner than he previously expected, possibly within just one or two years.

At the opening ceremony of the 26th China International Industry Fair, held from Oct 12 to 16 in Shanghai, Wang said his biggest mistake over the past 10 years or so was "not believing in AI enough".

Wang said that new advances in AI models such as GPT-6 could bring this major change within one or two years. He also said OpenAI's GPT-7, which he expects to be released next year, might completely solve the problems that are currently holding back AI-powered robots.

According to Shanghai Securities News, Wang said AI technology is improving very quickly around the world, with important updates coming every month or two. He said the recent release of GPT-6 has especially helped the robotics industry understand how AI can work with physical machines.

Wang said his company is testing this technology on many different kinds of robots and giving them different jobs to do. So far, the results have been very good. Unitree and other robot companies are already testing robots in factories. However, before robots can be widely used, they must become better at handling different tasks and working more efficiently. Wang believes these problems will very likely be solved within the next few years.

The phrase "ChatGPT moment" came after the phrase "iPhone moment." Nvidia CEO Jensen Huang used "iPhone moment" in 2023 when he compared the arrival of ChatGPT to the impact of the iPhone. Since then, people have used the phrase to describe a major turning point when a new technology or product goes from being an idea to something many people use. This kind of change can greatly affect an industry and how people use technology.

Wang's latest prediction is more hopeful than what he said earlier at the 2026 World Robot Conference in Beijing in August. At that time, he was less certain about how quickly AI-powered robots could reach such a major turning point.

On Aug 20, Wang said AI-powered robots might reach a major turning point like ChatGPT in as little as two to three years. But he also said it could take five to 10 years if progress was slower. One day earlier, Unitree had made its first appearance on the Shanghai Stock Exchange's science and technology innovation board, known as the STAR Market. Its shares closed 460 percent higher that day.

Wang has a clear way to decide when AI-powered robots have reached their "ChatGPT moment." He believes robots should be able to successfully finish about 80 percent of tasks in 80 percent of places they have never been before, simply by following spoken or written instructions.

Wang believes the biggest problem is making sure AI models can correctly understand information from physical robots and turn their decisions into actions the robots can perform. Unitree is continuing to test its robots in real-life situations. Some robots are already working in car factories, while the company's AI team is testing what they can do inside homes. However, Wang said the robots are not yet being used on a large scale because they still cannot work efficiently enough or handle enough different kinds of jobs.

At the opening ceremony of the 26th China International Industry Fair on Oct 12, Chen Jie, vice-president of the Chinese Academy of Engineering, said AI is an important technology that is helping bring major changes to technology and industry. He said AI will continue to help engineering and technology move forward.

Chen believes there is still a lot of work to do to understand how AI can help engineers create new technologies and improve existing ones. This includes finding useful ways to apply AI in different real-life situations, developing completely new ideas, and helping people and organizations work together. He also believes AI needs to become a much deeper and more complete part of engineering and technology to support the development of new industries and modern manufacturing.