Podcast Atlas
concept Updated 2026-08-07 Tags: Physical-Ai, Robotics, Embodied-Ai, Systems

Physical Intelligence System Stack

Physical intelligence system stack is Yao Song / 姚颂’s core definition of physical intelligence in 173: 对话姚颂：深鉴、东方空间、再出发，「天才少年」十年后. He argues that the field cannot be reduced to a single model, a robot body, or a visually impressive demo. It must combine data, compute, models, algorithms, software, hardware, scene access, solutions, and remote systems.

The concept extends Physical AI and Embodied AI Value Chain. Where many robotics discussions ask whether value sits in the robot brain, body, data, or scene owner, Yao’s version says early physical-intelligence companies may need to own enough of the full stack to make research, product, deployment, and business value move together.

Key Claims

  • Space perception, body motion, manipulation, and remote operation are all parts of physical intelligence rather than separate afterthoughts.
  • Vision Language Action Models, World Models, World Action Models, and Robot Reinforcement Learning are route components, not complete company definitions by themselves.
  • Scenario access matters because physical-world data and commercial value appear only when robots act in real settings.
  • Early vertical integration can be rational when standard interfaces, mature suppliers, and stable form factors have not yet formed.
  • The stack view supports Milestone Commercialization: each technical step should be linked to a deployable service or customer value where possible.

Connections