Skild AI, a robotics startup backed by Nvidia and SoftBank, is bringing its advanced artificial intelligence model to real-world manufacturing by deploying it on Foxconn’s assembly lines in Houston, where Nvidia’s Blackwell GPU server racks are produced. This marks one of the first commercial uses of what the companies describe as generalized physical AI.
The company’s technology, often referred to as a general-purpose “robot brain,” is designed to overcome a key limitation of traditional industrial robots, which are typically programmed for a single repetitive task. Skild AI’s model enables robots to adapt to multiple functions without requiring extensive reprogramming for each new process.
Explaining the scalability of the approach, CEO Deepak Pathak said, “If we partner with robotic OEMs (original equipment manufacturers) that already have hundreds of thousands of robots deployed, it gives us a path to extreme scalability and establishes the data flywheel.”
To expand its reach, Skild AI is also collaborating with major robotics players such as ABB Robotics and Universal Robots, integrating its software into widely used industrial machines. These partnerships aim to generate large volumes of operational data, which will help train and improve the AI system over time.
The deployment highlights a broader shift in manufacturing, where companies are moving toward more autonomous and flexible systems powered by AI. Unlike conventional automation, which relies on fixed programming, this new approach focuses on machines that can learn, adapt, and handle complex, variable tasks across different environments.
This development also aligns with growing investments in advanced manufacturing and AI infrastructure, as companies increasingly rely on automation to scale production efficiently. By integrating a universal AI layer into robotics, Skild AI and Nvidia are positioning themselves at the forefront of the next phase of industrial innovation.
