Skild AI Shows Robot Trained 140 Years in Simulation, Paxton on Sim-to-Learn Loops
chris_j_paxton · x · 2026-09-25
Robotics company Skild AI revealed its model was 'born in a physics simulation,' playing against increasingly tough previous versions of itself for the equivalent of 140 years. AI researcher Chris Paxton commented that as agents need to learn more complex and reactive behaviors, some kind of simulation will likely be required to drive the learning loop — a notable take on sim-based training for embodied intelligence.
More from Embodied
- Guava distills GPT-5.4 into a 4B robot agent with just 2,000 simulated trajectories — jbhuang0604 · 2026-11-09
- Dex-One2Many learns dexterous manipulation from one human video, 95%+ success across 4 hands — jbhuang0604 · 2026-11-09
- Fly-by-Code: coding agents pilot drones via active visual and physical feedback loops — jbhuang0604 · 2026-11-09
- μ0 world model learns robot physics from video only, beats π0 by 5 points on RoboCasa365 — jbhuang0604 · 2026-11-09
- NVIDIA unveils GR00T 2 open-world robot action model, hitting 95% success with 5 demos — CyberRobooo · 2026-11-09
- CoRL keynote to explore open-source data and hardware for robotics — micoolcho · 2026-11-09