NVIDIA announced collaborations with Korean industry leaders to expand the country’s AI infrastructure by deploying over 500M NVIDIA GPUs.
NVIDIA announced today that it is collaborating with South Korea to expand the country’s AI infrastructure by deploying over 250,000 NVIDIA GPUs across its clouds and AI factories. This infrastructure, developed through collaborative efforts between the public and private sectors, will support AI-driven economic growth and innovation across various industries, including automotive, manufacturing, and telecommunications.
Jensen Huang, founder and CEO of NVIDIA, highlighted Korea’s important role in the AI industrial revolution. He noted the need for a strong computing infrastructure that is as essential as power and broadband. He pointed out Korea’s long history of innovation across various sectors, including shipbuilding, automotive, semiconductors, and electronics. Furthermore, he emphasized that Korea has the potential to lead by producing “intelligence” as a new export, which could drive global change.
As world leaders gather in South Korea for the APEC Summit, the Ministry of Science and ICT (MSIT) has announced it is accelerating its plans to deploy up to 50,000 of the latest NVIDIA GPUs. This will boost AI development for enterprises and industries. The deployment will expand over the coming years, beginning with an initial rollout of 13,000 NVIDIA Blackwell and other GPUs. This effort involves NVIDIA Cloud Partner NAVER Cloud, along with NHN Cloud and Kakao Corp., to build up computing infrastructure on the country’s clouds through initiatives such as the establishment of Korea’s National AI Computing Center.
Research institutes, startups, and AI firms will utilize this infrastructure to develop models and applications, thereby supporting Korea’s national plan to enhance AI capabilities and infrastructure.
Additionally, NVIDIA is teaming up with industries, universities, and research institutions in Korea on AI-RAN and 6G infrastructure projects. It is collaborating with major companies, including Samsung, SK Telecom, ETRI, KT, LGU+, and Yonsei University, to develop smart, low-power AI-RAN network technology. This tech can reduce computing costs and extend device battery life by moving GPU computing tasks to the network’s base station.
Korea’s Industry Titans Build NVIDIA AI Factories for Advanced Manufacturing
Korea’s top industries are investing heavily in AI infrastructure to accelerate their enterprise and AI development.
Samsung is building a semiconductor AI factory that will use over 50,000 GPUs. This factory aims to enhance intelligent manufacturing and improve its products and services with AI. Samsung is using NVIDIA technologies like NVIDIA Nemotron post-training datasets, NVIDIA CUDA-X, the NVIDIA cuLitho library, and NVIDIA Omniverse to create digital twins that speed up advanced semiconductor manufacturing. The company is also leveraging NVIDIA Cosmos, NVIDIA Isaac Sim, and NVIDIA Isaac Lab to bolster its home robot development portfolio.
SK Group is establishing an AI factory capable of housing over 50,000 NVIDIA GPUs. This factory will support semiconductor research, development, and manufacturing, as well as cloud infrastructure for digital twin and AI agent development. SK Telecom plans to provide infrastructure with NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, enabling domestic manufacturers to access NVIDIA Omniverse. They will also offer industrial cloud infrastructure to speed up digital twin and robotics projects for startups, businesses, and government agencies.
Hyundai Motor Group and NVIDIA are deepening their collaboration. They will co-develop AI technology for mobility, smart factories, and on-device semiconductors using 50,000 Blackwell GPUs for AI model training and deployment. To support the Korean government’s goal of building a national AI cluster, Hyundai Motor Group and NVIDIA will work with government stakeholders to accelerate ecosystem development. This partnership is expected to see approximately $3 billion invested in improving Korea’s AI landscape. Important initiatives include establishing an NVIDIA AI Technology Center, the Hyundai Motor Group Physical AI Application Center, and regional AI data centers.
NAVER Cloud is also growing its NVIDIA AI infrastructure and plans to deploy over 60,000 GPUs, including NVIDIA RTX PRO 6000 Blackwell and other NVIDIA Blackwell GPUs, for sovereign and physical AI. NAVER Cloud is gearing up for the next phase of sovereign AI development in Korea, powered by NVIDIA Nemotron open models running on its NVIDIA AI infrastructure. NAVER Cloud aims to create industry-specific AI models, such as those for shipbuilding and security, focusing on inclusive AI for Korea’s citizens.
Korea Government and Developers Advance LLM Research With NVIDIA
Korea’s MSIT, in partnership with NVIDIA, is making progress on its Sovereign AI Foundation Models project. This initiative aims to develop sovereign language models. The project will use NVIDIA NeMo and open NVIDIA Nemotron datasets to leverage local data for building and refining reasoning models.
LG AI Research, NAVER Cloud, NC AI, SK Telecom, and Upstage are collaborating to develop sovereign models. Enterprises, researchers, and startups will be able to contribute to and utilize these models for creating AI agents with speech, reasoning, and other abilities.
LG is partnering with NVIDIA to support the growth of physical AI technology and enhance the physical AI ecosystem. They are also working to assist startups and academia with LG’s EXAONE models, including the EXAONE Path healthcare model, which was developed using the MONAI framework to aid cancer diagnosis.
Korea and NVIDIA Advance Quantum Computing and Scientific Research
The Korea Institute of Science and Technology Information (KISTI) is collaborating with NVIDIA to set up a Center of Excellence for research using Korea’s sixth-generation national supercomputer, HANGANG, which features NVIDIA-accelerated computing.
KISTI has also announced support for the new NVIDIA NVQLink open architecture, which connects quantum processors and GPU supercomputing. By using the NVIDIA CUDA-Q platform, NVQLink will help KISTI advance research into areas like quantum error correction and hybrid application development, speeding up the development of future quantum-GPU supercomputers.
KISTI will also create foundation models for scientific research and help researchers develop physics-informed AI models using the open-source NVIDIA PhysicsNeMo framework.
New Startup Alliance Supports Korean Development
To encourage economic growth and opportunities in Korea, NVIDIA and its partners are forming an alliance to support startups through the NVIDIA Inception program. Members will have access to advanced computing resources from NVIDIA Cloud Partners, including SK Telecom, alongside support from NVIDIA Inception and VC Alliance members like IMM Investment, Korea Investment Partners, and SBVA. Startups will benefit from NVIDIA’s software and expertise, helping them accelerate the growth of the next generation of companies.
Building on its efforts through the NVIDIA Inception program for startups, NVIDIA also plans to collaborate with the Korean government to support future companies. It will participate in the N-Up AI startup incubation program run by the Korea Ministry of SMEs and Startups.
Engage with StorageReview
Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed
