LG and Nvidia Join Forces on Humanoid Robots, AI Factories, and Autonomous Vehicles

From a technical standpoint, the robot will run on Nvidia's Isaac GR00T — the company's open humanoid foundation model and development platform — and will be equipped with Nvidia's Jetson Thor GPUs for onboard computing, providing advanced reasoning and control at the edge . The robot will also incorporate Nvidia Halos, billed as the industry's first full-stack safety framework for robots .

LG's role in the joint development leverages its sprawling manufacturing ecosystem: LG Electronics will supply actuators, LG Innotek will provide sensors, and LG Energy Solution will supply batteries . This pooling of affiliate technologies is a hallmark of how LG Group intends to combine Nvidia's AI brain with its own hardware muscle .

Before the bipedal robot arrives, LG will deploy CLOiD wheel-based service robots at LG Electronics' washing-machine production line in Tennessee as a proof of concept, expected by the end of 2026 . This real-world deployment on a live factory floor will help validate the integration of Nvidia's AI platforms with LG's hardware .

AI Factory Infrastructure: 80 MW Facility in Cheonan

The partnership extends well beyond robots into large-scale AI infrastructure. LG is building an 80-megawatt AI factory in Cheonan, South Korea, with a target completion year of 2028 . This facility will integrate Nvidia's AI computing platforms — specifically, the companies plan to build an Nvidia Vera Rubin-based AI factory reference site in the first half of 2027, which will later scale into the full Cheonan facility .

The MoU also includes technical cooperation on a "robot data factory" based on LG CNS's robotics platform, PhysicalWorks. This system is designed to repeatedly collect and generate data, train models, and verify results — a closed loop that accelerates the development cycle for physical AI .

Autonomous Vehicle Technology: DRIVE Hyperion Platform

On the mobility front, the two companies will collaborate on next-generation vehicle computing based on Nvidia's DRIVE Hyperion platform . LG will contribute its automotive component expertise, including vehicle domain controllers, cameras, sensors, and communications modules, to develop an AI-Defined Vehicle (ADV) platform . This builds on LG's existing electronics and automotive parts business, extending the partnership into the software-defined vehicle space.

Competitive Context: Taking on Tesla and Unitree

The LG-Nvidia MoU is widely seen as a direct challenge to Tesla's Optimus humanoid robot program and China's Unitree, as well as other players like Figure AI and Agility Robotics . By combining Nvidia's dominant AI computing stack with LG's vast manufacturing scale and global supply chains, the two companies aim to set de facto industry standards for physical AI before competing ecosystems mature . A joint task force has been established to accelerate R&D through on-site demonstrations and commercialization .