Reward AI Releases OM-1: A Robot Policy Trained on Human Demonstrations Only, With No Teleoperation or On-Robot Data
AITopTools Editorial TeamSeptember 14, 2026
What changed
Reward AI has released OM-1, a robot-control system trained only on human demonstrations recorded with a wearable glove, rather than remote operation or data collected by robots. The company says it works with industrial robot arms and humanoids at human speed and can learn a new task from less than 30 minutes of examples.
What this means for you
This could eventually make it easier to teach different kinds of robots by showing them how to perform a task. However, OM-1 is not yet publicly available: its software, model files, and programming interface have not been released.