Nvidia to supply 260,000 chips to Korea for AI factories with Samsung, SK, Hyundai

President Lee Jae Myung poses with Nvidia CEO Jensen Huang, third from left, and Korean business leaders during the Asia-Pacific Economic Cooperation summit in Gyeongju, North Gyeongsang Province, Friday. Nvidia announced plans to deploy up to 260,000 graphics processing units to Korea in partnership with the government and major companies to build large-scale artificial intelligence factories. From left are Naver founder Lee Hae-jin, SK Group Chairman Chey Tae-won, Huang, Lee, Samsung Electronics Executive Chairman Lee Jae-yong and Hyundai Motor Group Executive Chair Chung Euisun. Yonhap

GYEONGJU, North Gyeongsang Province — Nvidia has unveiled a landmark collaboration with Korea’s leading conglomerates — Samsung, SK Group and Hyundai Motor Group — to build artificial intelligence (AI) factories. The company announced it will supply a total of 260,000 Nvidia Blackwell graphics processing units (GPUs) across these three companies as well as Naver and the Korean government.

President Lee Jae Myung said the large-scale partnerships mean Nvidia has started investing in Korea’s AI industry.

“AI is also a new industry in itself because it needs manufacturing, factories, deep technical expertise, innovative and visionary entrepreneurs, as well as industrial factory capability. These are the perfect ingredients and capabilities for Korea to fly,” Nvidia CEO Jensen Huang said during a meeting with the president and heads of Samsung, SK, Hyundai and Naver on the sidelines of the Asia-Pacific Economic Cooperation (APEC) summit in Gyeongju, North Gyeongsang Province, Friday.

“Korea has deep technical capabilities, visionary entrepreneurs and, of course, no country in the world is better at industrial capability than Korea.”

Nvidia will deploy 50,000 GPUs to Samsung to build a new AI megafactory that integrates the entire semiconductor manufacturing process into an AI-powered platform to optimize development and production in real time.

Leveraging Nvidia’s cuLitho, CUDA-X and Omniverse libraries, Samsung will build digital twins to improve the speed and yield of semiconductor manufacturing while developing next-generation home robots using Nvidia Cosmos and Isaac GR00T.

Nvidia will also supply 50,000 GPUs to SK Group to build an AI cloud for manufacturing, which will incorporate Nvidia’s Omniverse to utilize digital twin technology for efficient and cost-effective semiconductor manufacturing. SK Telecom plans to provide sovereign infrastructure featuring Nvidia RTX Pro 6000 Blackwell server edition GPUs, enabling domestic manufacturers to leverage Omniverse.

Nvidia CEO Jensen Huang delivers a keynote speech at a session of the Asia-Pacific Economic Cooperation CEO Summit in Gyeongju, North Gyeongsang Province, Friday. Joint Press Corps

With Hyundai Motor Group, Nvidia will power the automotive conglomerate’s new AI factory with 50,000 Blackwell GPUs to accelerate innovations in AI-driven mobility. The two companies will co-develop AI capabilities for mobility solutions, next-generation smart factories and on-device semiconductor advancements.

The two companies will also jointly invest approximately $3 billion in advancing Korea’s physical AI landscape, signing a trilateral partnership agreement with the Ministry of Science and ICT on Friday in Gyeongju. As part of the efforts, they will establish key AI institutions in the country, including the Nvidia AI Technology Center, Hyundai’s Physical AI Application Center and physical AI data centers.

Nvidia will additionally supply 60,000 GPUs to Naver Cloud and 50,000 GPUs to the Korean government. Naver announced that it had signed a memorandum of understanding with Nvidia to jointly develop a next-generation physical AI platform and build AI infrastructure to support applications across major industries, including semiconductors, shipbuilding and energy.

Nvidia’s Nemotron open models will be utilized in the Korean government’s national AI foundation model project to develop Korea’s proprietary AI models.

Nvidia CEO Jensen Huang speaks during the Live Keynote Pregame during the Nvidia GTC artificial intelligence conference in Washington, D.C., Tuesday (local time). AFP-Yonhap

During the meeting with Huang, Lee expressed hopes that the company will join Korea’s national drive to become one of the world’s top three AI powerhouses, highlighting the government's commitment to innovation.

“Korea aims to become the AI capital of the Asia-Pacific region, with global companies such as BlackRock and OpenAI joining the project,” he said.

“I hope Nvidia would also take part in building a virtuous AI ecosystem where infrastructure, technology and investment reinforce one another.”

The government plans to work with Nvidia to broaden startup support through the N-Up AI startup incubation program, operated by the Ministry of SMEs and Startups.

“While Nvidia drives the pace of AI innovation, Korea stands as the ideal partner to channel that momentum and guide it in the right direction,” the president said.

“The Korean government pledges full support to ensure that the partnership discussed today becomes a successful case contributing not only to Korea but also to the global community.”

In the telecom sector, Nvidia will work with Samsung, SK Telecom, KT, LG UPlus, Yonsei University and the Electronics and Telecommunications Research Institute to develop and commercialize next-generation wireless technologies, AI radio access networks and AI-native 6G.

Nvidia and the Korea Institute of Science and Technology Information (KISTI) will also advance quantum computing research, leveraging Korea’s sixth national supercomputer, Hangang, which is powered by Nvidia. The two will jointly research hybrid quantum computing using Nvidia’s quantum processing unit-agnostic platform, CUDA-Q, and build a foundation model for scientific research and development. KISTI will also train its researchers in developing AI models using Nvidia’s open-source PhysicsNeMo framework.

Additionally, Nvidia is working with LG to advance robotics and support startups and academia with LG’s EXAONE models, including applications for cancer diagnosis.