NVIDIA launched GR00T 2, a world action model built on NVIDIA Cosmos that predicts future video and robot actions together, at the Conference on Robot Learning (CoRL 2026) in Austin

Systems trained on data to learn patterns and make predictions or generate outputs, such as text, images, or decisions, by approximating relationships between inputs and outputs without being explicitly programmed for each task.

The fusion of artificial intelligence and mechanical engineering that enables machines to perceive their environment, make decisions, and perform physical tasks autonomously.