rewrite this title NVIDIA Strengthens AI Factory Strategy With SK hynix Deal And Expanded South Korean Data Center Initiatives | Metaverse Post

by
Alisa Davidson

Published: June 08, 2026 at 3:14 am Updated: June 08, 2026 at 3:14 am

by Anastasiia O

Edited and fact-checked:
June 08, 2026 at 3:14 am

To improve your local-language experience, sometimes we employ an auto-translation plugin. Please note auto-translation may not be accurate, so read original article for precise information.

Technology company NVIDIA announced a new collaboration with South Korean semiconductor manufacturer SK hynix aimed at advancing next-generation memory technologies to support the expansion of global AI infrastructure and accelerate innovation in semiconductor design and production. The agreement builds upon a longstanding engineering relationship between the two companies that has contributed to the development of advanced AI computing systems over several years.

The multi-year arrangement is intended to strengthen memory supply planning for future technologies, reflecting the lengthy development timelines associated with advanced memory products. As demand for AI computing capacity continues to grow worldwide, the partnership is designed to align memory development with NVIDIA’s long-term infrastructure plans and the broader expansion of AI-focused data centers and computing facilities. The companies will work together on memory technologies for a range of NVIDIA platforms, including Vera Rubin AI supercomputers, Vera processors, RTX Spark-powered personal computing systems, and Jetson Thor robotics platforms. The cooperation also reflects the emergence of new markets linked to AI infrastructure, personal AI applications, and physical AI systems.

Beyond memory development, SK hynix is incorporating NVIDIA’s CUDA-X software libraries and AI technologies into semiconductor simulation processes. These tools are being applied to areas such as technology computer-aided design and computational lithography, with the goal of improving the speed and efficiency of research and development activities. The company is also using the NVIDIA PhysicsNeMo framework alongside CUDA-X to accelerate internal simulation workloads and physics-based AI modeling. Industry observers note that broader adoption of such technologies could encourage closer collaboration among semiconductor manufacturers, NVIDIA, and providers of electronic design automation software.

The partnership further extends to manufacturing innovation through the development of digital twin technologies for semiconductor fabrication facilities. SK hynix is creating virtual factory environments that can model and simulate complex production operations. By utilizing NVIDIA Omniverse technologies and OpenUSD-based workflows, engineers can generate detailed three-dimensional representations of manufacturing sites to analyze performance, test operational scenarios, and identify opportunities for improvement before changes are implemented in physical facilities.

These digital twin systems may also contribute to operational efficiency by supporting the management of autonomous mobile robots and other factory assets. Technologies such as the GPU-accelerated NVIDIA cuOpt optimization engine and the NVIDIA Metropolis platform are being evaluated for their potential to enhance logistics, scheduling, and resource allocation within semiconductor plants. The companies are additionally exploring methods for integrating digital twin environments with existing software systems and AI-driven workflows, enabling automated analysis of manufacturing data and assisting decision-making processes across production operations.

The collaboration reflects the growing importance of memory technology within the AI industry. Training and operating advanced AI models require the movement and processing of vast quantities of data, making memory performance a critical factor in overall system efficiency. High-bandwidth and high-capacity memory solutions play an increasingly important role in ensuring that AI processors can operate at their full potential while managing power consumption and performance requirements.

NVIDIA Expands South Korea AI Infrastructure Push Through New Strategic Partnerships

In addition to its collaboration with SK hynix, NVIDIA announced several other partnerships aimed at supporting the expansion of AI infrastructure across South Korea. Among the initiatives unveiled, SK Telecom revealed plans to develop a gigawatt-scale AI cloud platform powered by NVIDIA technologies. The project is expected to include the construction of an AI-focused data center scheduled to begin operations in 2027, reflecting growing investment in large-scale computing infrastructure to meet rising demand for artificial intelligence services.

NVIDIA also disclosed that South Korean technology company Naver and industrial conglomerate Doosan would adopt its technologies as part of their efforts to develop AI data center capabilities. The collaborations are intended to support the deployment of advanced computing environments designed to handle increasingly complex AI workloads.

Doosan indicated that it sees opportunities to integrate its energy-related solutions into future NVIDIA data center platforms. The company, which is involved in robotics development and supplies materials used in NVIDIA’s advanced Blackwell processors, stated that it is also evaluating the use of NVIDIA’s physical AI technologies in support of its broader industrial and automation initiatives.

The U.S. chipmaker further expanded its cooperation with South Korea’s LG Group. According to NVIDIA Chief Executive Jensen Huang, the companies are working together across a range of areas that include electronics, mechanical engineering systems, and artificial intelligence technologies for humanoid robotics applications. The announcement followed discussions with LG Group Chairman Koo Kwang-mo regarding future areas of technological cooperation.

Beyond robotics, the two companies are also collaborating on next-generation data center infrastructure. Their efforts include the design of key architectural elements such as cooling systems, power distribution networks, and integrated facility layouts. The partnership is intended to address the growing technical requirements of AI computing facilities, with both companies exploring approaches that encompass the full lifecycle of data center planning, construction, and operation.

Alisa, a dedicated journalist at the MPost, specializes in crypto, AI, investments, and the expansive realm of Web3. With a keen eye for emerging trends and technologies, she delivers comprehensive coverage to inform and engage readers in the ever-evolving landscape of digital finance.

More articles