[REVISION HISTORY]

South Korean electronics firms expand robotics and AI

16 recorded revisions of this situation.

What changed

2026-09-30 14:23 UTC → 2026-10-21 02:54 UTC · added ~~removed~~

In July 2026, LG Electronics established a dedicated Robotics Business Unit to commercialize robotics across industrial, commercial, and residential markets. A central component is the specialized Data Factory at the Yangjae R&D campus in Seoul, designed to generate operational data for LG’s Robot Foundation Model (RFM). By the end of 2026, this facility is expected to house several hundred robots to produce approximately 100,000 hours of training data using NVIDIA technologies like Omniverse and Isaac to enhance robot learning for tasks such as assembly and cleaning. Concurrently, Samsung Electronics launched its RX (Robotics eXperience) division and increased its stake in Rainbow Robotics to approximately 35%, aiming for AI-autonomous manufacturing by 2030. ~~Following a memorandum of understanding signed in Santa Clara,~~ LG has deepened its partnership with ~~Nvidia~~ NVIDIA to integrate ~~LG’s~~ hardware with ~~Nvidia’s~~ NVIDIA’s AI ~~stack. This collaboration focuses~~ stack, focusing on physical AI, AI infrastructure, and mobility. This includes developing next-generation AI-driven vehicle (AIDV) platforms by combining LG’s software with NVIDIA’s Drive Hyperion. LG is also developing a bipedal humanoid robot utilizing ~~Nvidia’s~~ NVIDIA’s Isaac GR00T ~~foundation model~~ and Jetson ~~Thor platform,~~ Thor, with a planned unveiling by Q1 2027. To support these goals, LG plans to deploy CLOiD robots in a Tennessee washing machine factory by late 2026 and aims to complete an 80-megawatt AI factory in Cheonan, South Korea, by the first ~~quarter~~ half of ~~2027.~~ 2028. In September 2026, LG hosted ‘LG SPARK 2026’, an AI innovation festival. During the event, LG AI Research showcased its ‘EXAONE’ model family, highlighting specialized ‘Expert AI’ solutions for manufacturing, finance, and science. Notable achievements included the ‘EXAONE Discovery’ model, which identified a hair loss treatment candidate in one day—a process that previously took 22 months. In manufacturing, the ‘EXAONE Tabular’ and ‘EXAONE Omni-Inspect’ models contributed to an 85% reduction in product lead times at LG Innotek. Expanding its infrastructure footprint, LG Group entered a partnership with Microsoft to support global AI data center (AIDC) expansion. By late September 2026, LG Group companies secured qualifications for NVIDIA’s DSX Ready program. LG Electronics has since joined the NVIDIA Partner Network (NPN) as a Power and Cooling Solution Preferred Partner. Its 2.6MW Coolant Distribution Unit (CDU) has qualified as NVIDIA DSX Ready, representing the largest cooling capacity among current DSX Ready CDU partners.

Versions

Select a version to see what changed compared to the one before it.