NVIDIA has announced a major expansion of its AI infrastructure partnerships in South Korea, marking a pivotal moment in the country’s push to become a global leader in AI-driven industry transformation. At the APEC Summit 2025, NVIDIA revealed plans to collaborate with SK Group, Samsung Electronics, Hyundai Motor Group, and Naver to build a new generation of AI factories and cloud platforms, backed by over 260,000 of its latest GPUs. Of these, 50,000 will be dedicated to public initiatives, including the development of sovereign AI foundation models and a national AI data center, while the rest will support private sector innovation. A cornerstone of the initiative is the construction of a massive AI factory by SK Group, featuring more than 50,000 NVIDIA GPUs, with the first phase set for completion by late 2027. This facility will serve SK subsidiaries like SK hynix and SK Telecom (SKT), as well as external organizations through a GPU-as-a-service model. The infrastructure will power digital twins, robotics, and AI agent development, enabling real-time simulation, optimization, and autonomous decision-making across manufacturing and industrial systems. SK hynix is already using NVIDIA’s CUDA-X and PhysicsNeMo frameworks to accelerate chip design with AI physics, while deploying digital twins via NVIDIA Omniverse to improve fab efficiency and production ramp-up. SKT is also building an industrial AI cloud powered by over 2,000 NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, supporting advanced workloads in semiconductor manufacturing, robotics, and AI agent development. Meanwhile, Naver Cloud is launching a next-generation “Physical AI” platform, connecting real-world industrial systems with digital environments to drive innovation in semiconductors, shipbuilding, energy, and biotechnology. Samsung is taking a leading role with its own AI Megafactory, also powered by over 50,000 NVIDIA GPUs. This facility will integrate AI into every stage of semiconductor, mobile, and robotics production, using NVIDIA Omniverse for digital twins and cuLitho for 20x faster computational lithography. The collaboration with Samsung extends beyond hardware—NVIDIA and Samsung are co-developing HBM4 memory and advancing AI-RAN (AI-powered radio access networks) with Korean telecom operators and ETRI, aiming to create a global testbed for next-generation 6G infrastructure. Hyundai Motor Group is joining the effort to build AI-powered smart factories and autonomous mobility systems, deploying 50,000 Blackwell GPUs for AI model training and research. The partnership will focus on software-defined vehicles, robotics, and intelligent manufacturing, with joint AI research centers in South Korea. These initiatives underscore a broader trend: the fusion of AI, hardware, and physical systems. NVIDIA’s CEO Jensen Huang hailed the partnerships as a new era of “AI industrial revolution,” where intelligent computing transforms design, manufacturing, and mobility. The collaboration, spanning 25 years of partnership with Samsung and now deepening with SK, Hyundai, and Naver, positions South Korea as a global hub for physical AI and next-generation semiconductor innovation. As NVIDIA’s market cap surpasses $5 trillion, these strategic alliances signal a new phase in the global AI race—where infrastructure, industry, and government converge to build the future.