How to AI-generate a robot in the physical world | Tsun-Hsuan Johnson | TEDxBoston

Artificial intelligence has grown tremendously in recent years. What if, beyond serving as a smarter brain, it could create its own body to interact with the physical world? This talk introduces how we can use generative AI to build soft robots. Previously, generative AI excelled in digital content creation like writing emails and making graphics. However, the lack of physical knowledge has hindered its ability to create designs with the physical utility required for robotics. We demonstrate how to leverage physics-based simulations on top of generative AI to design various robots capable of performing tasks in the physical world.

3D Technology, Computers, Engineering, Intelligence, Research, Robots Tsun-Hsuan Johnson Wang is a PhD student in the Distrbuted Robotics Lab at MIT CSAIL. He received his B.Sc. and M.Sc. at National Tsing Hua University (NTHU) in Taiwan. He also did research in the industrial research groups such as Liquid AI, MIT-IBM Watson AI Lab, Uber Advanced Technologies Group, etc. His research lies in the intersection of robotics, simulation, and machine learning and his goal is to build intelligence that can interact reliably with the physical world. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at https://www.ted.com/tedx