35: NVIDIA’s Rev Lebaredian on Building Robot Brains and the Promise of Physical Artificial Intelligence

Building general-purpose "robot brains" requires moving beyond traditional, hard-coded programming toward generative AI and large language models that allow robots to perceive, plan, and act in the physical world. This transition relies on "AI factories"—specialized computing centers that manufacture intelligence by processing massive datasets. Because real-world data is limited, simulation technology and digital twins are essential for training robots in virtual environments that mirror physical laws. By creating high-fidelity digital representations of factories and workspaces, developers can test millions of configurations to optimize safety and efficiency before physical deployment. Humanoid form factors are increasingly prioritized to ensure these robots can effectively operate within environments already designed for human anatomy. Rev Lebaredian, Vice President of Omniverse and Simulation Technology at NVIDIA, emphasizes that this integration of computing and physical reality represents the next major frontier in industrial innovation.