Nvidia Scientist Predicts 'GPT-3 Moment' for Robotics in Next Few Years
TapTechNews September 19th news, the tech media TheDecoder released a blog post yesterday (September 18th), reporting that Nvidia senior scientist Jim Fan predicted that the field of robotics will usher in a 'GPT-3 moment' in the next few years.
Jim Fan Profile
Jim Fan obtained his Ph.D. from the Vision Laboratory at Stanford University, under the tutelage of Professor Li Feifei. His research fields are very extensive, including multi-modal basic models, reinforcement learning, and computer vision. He has interned at well-known organizations such as Google Cloud AI, OpenAI, and Baidu Silicon Valley AI Laboratory.
Jim Fan is currently leading AI-related research at Nvidia, and his team is committed to developing 'Project Groot', which is the company's effort to create a basic model for humanoid robots.
Research Breakthroughs in the Next Two to Three Years
Jim Fan predicted that within the next 2-3 years, there will be significant breakthroughs in the research related to the basic model of robots, but he also admitted that it will take a longer time for robots to enter daily life.
When interviewed by Sequoia Capital, Fan said he expects a 'GPT-3 moment' in the field of robotics - that is, a breakthrough in the basic robot model, and its influence can be comparable to the impact of GPT-3 in the field of language processing.
TapTechNews translated his views as follows:
To let robots enter people's daily lives is not just a technical issue. Robots need to be affordable and capable of mass production, and at the same time, we also need hardware security and privacy and regulatory guarantees.
The world is built around the human form, right? Our restaurants, factories, hospitals, and all equipment and tools - they are all designed for the human form and hands.
He believes that, in theory, a highly capable humanoid robot can perform any task that a human can accomplish, and predicts that the ecosystem of humanoid robot hardware will be ready in two to three years.
Nvidia's Robotics-Related Research
Nvidia uses a combination of three types of data when developing robot AI: Internet data, simulation data, and real-world robot data. Dr. Fan emphasized the advantages and disadvantages of each method and believes that their combination is the key to success.
Nvidia is developing technologies such as 'Eureka', which uses language models to generate reward functions for robot training to achieve an automated process.
In addition to the real world, Fan's team is also researching AI agents for virtual environments such as video games. He finds similarities between these fields and is committed to developing a unified model that can control both virtual and physical agents in the long term.