RICE AI

RICE AI is a crowdsourced platform for robotics data collection. It allows teleoperators to remotely control robots and capture data such as robot vision, joint movements, and force measurements. Operators can use webcams with skeleton detection, joysticks, VR controllers, or teleoperation rigs—and earn token rewards based on the device’s dexterity.

Individuals and companies can also contribute robots—whether humanoids, mobile bimanual robots, quadrupeds, or AI toys—to dedicated data collection centers, receiving token rewards in return. Rewards scale higher for robot types with greater demand. These centers are located in both real-world business settings and lab-based environments.

The collected data is sold to research labs and companies. Additionally, RICE AI trains robotics foundation models using this data and provides access via a subscription service. Clients who pay with the native token receive discounts. Proceeds from data sales are partially used to burn tokens, reducing supply. The token also functions as a governance tool for the platform.

To ensure data quality, an AI evaluation model reviews teleoperator submissions, supported by human labelers who rank samples based on task efficiency. A similarity-prediction model and clustering-based embedding database further filter redundant data, ensuring only high-quality datasets remain.