Pre-training leverages alternate data sources, enabling Skild AI to build foundational capabilities prior to real-world deployment.

Skild AI is building a scalable foundation model for robotics intended to act as a shared brain across embodiments and real-world tasks, backed by data-driven simulations and a vast dataset.

Management signals plans to extend deployments from enterprise applications to consumer home robotics as a key growth axis.

Leadership says the company aims to accelerate deployments by scaling model training and expanding across diverse environments and tasks, positioning itself as a driver of automation in U.S. manufacturing and beyond.

CEO Deepak Pathak and Co-Founder Abhinav Gupta describe the Skild Brain as a continuous, data-flywheel-driven system focused on intelligence in physical environments rather than rote memorization.

The industry is seeing parallel efforts from Field AI and other players, signaling growing interest in universal robotic cognition and easily adaptable software.

The Skild Brain is trained on diverse data sources, including learning by watching human videos and physics-based simulations, to compensate for the lack of a robotics internet.

Skild AI, a robotics software maker founded in 2023, has secured a valuation above $14 billion after a $1.4 billion Series C led by SoftBank, with participation from Nvidia, Macquarie Group, and 1789 Capital among others.

The company emphasizes in-context learning as a core capability, enabling adaptations to new bodies or environments in real time without retraining, a step toward general intelligence in physical robots.

The firm touts a large-scale data-driven approach, claiming a dataset many times larger than competitors’ and heavy use of simulations to advance Physical AI.

Skild AI operates within a broader trend of learn-as-you-go robotic software and the rise of humanoid-focused AI robotics, addressing the training burden for widespread robot adoption.

Executives and investors stress the cross-robot, cross-environment potential of Skild AI and the strategic value of Physical AI for commercial and national interests.