Reward AI’s OM-1 Model: Why Human Motion Capture Is Accelerating Robot Training
US-based Reward AI has released videos showcasing its Omnibody Model 1 (OM-1) foundational AI model for robots, which learns complex tasks solely from human movements captured by a wearable device.
Dick Weisinger · September 19, 2026 · 2 min read · Source: ITmedia NEWS · Issue #101
East Asian Technology Intelligence
Japan & China tech news — translated, contextualized, and delivered for Western readers.
Free. Unsubscribe anytime.
This story ran in Issue #101, alongside three other stories.
Robotics & Automation
US-based Reward AI has released videos showcasing its Omnibody Model 1 (OM-1) foundational AI model for robots, which learns complex tasks solely from human movements captured by a wearable device. The videos, presented at 1x speed, show robots adeptly performing tasks like packing smartphones, mixing cocktails, folding laundry, and unplugging Ethernet cables, drawing widespread attention on X for their speed and fluidity. OM-1 is designed for cross-robot compatibility, from industrial arms to humanoids, and can learn new tasks from less than 30 minutes of human demonstration.
This development represents a step towards more intuitive and rapid robot programming, moving away from complex teleoperation or reinforcement learning on physical robots. For Western readers, it points to a potential acceleration in robot deployment across diverse industries, from manufacturing to service, by simplifying the teaching process.