NVIDIA's Robo Olympics uses Codex and natural language to train robot skills in physics simulation
RevLebaredian · x · 2026-09-30
NVIDIA Robotics showcased Robo Olympics, an exploratory project built by Tae Kim, VP of Engineering for NVIDIA Omniverse, using Codex with OpenAI's GPT-6 Astra model driven purely by natural-language instructions.
The core idea: simulators are time machines that let robots validate every move in physics simulation before touching the real world. The author argues AI agents can now automate simulation development that once required teams of specialists, drastically lowering the barrier to physical AI development.
Related event: NVIDIA Shows Codex and GPT Powering Automated Robot Simulation(2 posts)→
More from Embodied
- CrossBFM Distills a Shared Behavior Space Across Humanoid Robots in Under One GPU-Hour — Tan-Dzung Do · 2026-09-30
- Real2Gym Turns Videos Into Interactive Robot Gyms, Beating GPT-6 Direct Mode by 33% on Real Robots — Shanghai-AI-Laboratory · 2026-09-30
- Survey Maps Four Families of In-Context Learning for Robots, Toward Physical Recursive Self-Improvement — Knowin · 2026-09-30
- RobotGym launches free public MCP for Physical AI simulation with hosted GPUs — rocky_mountain12 · 2026-09-30
- USC's CLAM learns robot policies from unlabeled videos, 2-3x success over baselines — ebiyik_ · 2026-09-30
- A huggable humanoid robot called Baymax shows up at IROS 2026 — CyberRobooo · 2026-09-30