OpenAI trains robotic hand to solve a Rubik's Cube with Automatic Domain Randomization
Evidence notes
- OpenAI trained neural networks to solve a Rubik's Cube with a human-like robotic hand using reinforcement learning in simulation.
- The project introduced Automatic Domain Randomization to create progressively harder simulated environments for sim-to-real transfer.
- The system solved simpler scrambles 60 percent of the time and maximal-difficulty scrambles 20 percent of the time, making it a durable dexterous-manipulation milestone.
Company context
OpenAI develops foundation models, AI products and research systems, with a renewed robotics program focused on general-purpose robotics, robotic data acquisition, perception, simulation and model evaluation in physical environments. Its robotics relevance is model-layer and infrastructure-led rather than OEM-led.