NAVER, South Korea's tech giant, is partnering with NVIDIA to build massive AI factories that will serve both Korean industries and global customers, starting with a 55-megawatt deployment and scaling to gigawatt-level capacity. The initiative, powered by NVIDIA's DSX (Data Center for Sovereign AI) platform, represents a significant shift in how countries are approaching artificial intelligence infrastructure, moving beyond experimentation to production-scale AI systems that comply with regional data sovereignty requirements.

What Is NVIDIA's DSX Platform and Why Does It Matter?

NVIDIA's DSX platform is a unified AI factory architecture that combines hardware, software, facilities, and partner technologies designed to reduce the cost of generating AI responses and accelerate deployment timelines. Think of it as a complete blueprint for building AI infrastructure at scale, rather than piecing together separate components. The platform includes specialized software like DSX MaxLPS, which maximizes the number of AI responses processed per megawatt of power, and DSX OS, which handles lifecycle management, system reliability, automation, and supporting multiple AI services simultaneously.

NAVER will deploy this technology at its GAK Sejong Data Center in South Korea, a hyperscale facility designed to handle high-density accelerated computing with built-in automation, sustainability features, and disaster-response capabilities. The initial 55-megawatt deployment is just the beginning; the companies plan to scale this to gigawatt-level capacity, which would represent a massive expansion of AI computing resources.

How Will NAVER Use This AI Infrastructure?

• Language Model Development: NAVER is enhancing its HyperCLOVA X family of AI models by fine-tuning NVIDIA's Nemotron 3 Ultra open model using proprietary datasets and training expertise, resulting in models with stronger Korean-language capabilities and support for sovereign AI deployments across Europe and the Middle East.
• AI Agent Platform Launch: The company plans to launch an AI Agent Platform in Korea during the second half of 2026 using NVIDIA NemoClaw blueprints, enabling businesses to deploy AI agents that can perform tasks autonomously.
• Seoul World Model Creation: NAVER will develop a Seoul World Model based on proprietary urban street-view data and spatial modeling technology, creating a detailed digital representation of the city for various applications.
• Enterprise and Government Services: The infrastructure will support sovereign AI initiatives, enterprise AI workloads, government applications, and large-scale AI cloud services across multiple regions.

NAVER also became the first Korean company to join the NVIDIA Nemotron Coalition, contributing to open model development across pretraining, post-training, and reinforcement learning efforts. This positions the company as a key player in shaping how open-source AI models are developed globally.

Why Is Jensen Huang Emphasizing Sovereign AI Infrastructure?

"Useful AI has arrived, and demand for AI factories is extraordinary. NAVER is building AI factory infrastructure that will serve its companies, developers and industries. With NVIDIA DSX, we can help Korea scale sovereign intelligence infrastructure for the agentic era, from AI agents to AI factories and physical AI," said Jensen Huang, founder and CEO of NVIDIA.

Jensen Huang, Founder and CEO, NVIDIA

Huang's emphasis on "sovereign intelligence infrastructure" reflects a broader global trend. Countries and enterprises increasingly want to build AI systems that remain under their control and comply with local data protection laws, rather than relying entirely on foreign cloud providers. The "agentic era" Huang mentions refers to the next phase of AI development, where AI systems will operate more autonomously, making decisions and taking actions on behalf of users and organizations.

NAVER's founder and chairman, Haejin Lee, echoed this vision, stating that the company is building infrastructure capable of serving both Korean industries and global customers with trusted, high-performance AI. He emphasized that the DSX platform will help customers transition from experimenting with AI to deploying production-scale AI factories that power models, agents, and real-world services.

What Does This Mean for the Global AI Landscape?

The NAVER-NVIDIA partnership signals that the AI infrastructure market is becoming increasingly competitive and geographically distributed. Rather than concentrating AI computing power in a few Western data centers, countries are investing in their own sovereign AI infrastructure. This approach allows nations to maintain control over sensitive data, comply with local regulations, and build AI capabilities that serve their specific industries and languages.

The gigawatt-scale ambition is particularly significant. A gigawatt of power is enough to supply roughly 750,000 homes. Scaling NAVER's deployment to that level would make it one of the world's largest AI computing facilities, capable of training and running massive AI models for thousands of organizations simultaneously. This kind of infrastructure investment demonstrates how seriously major tech companies and nations are taking the AI transition.

The partnership also highlights NVIDIA's strategy of positioning its technology as the foundation for sovereign AI globally. By providing the DSX platform and working with companies like NAVER, NVIDIA is becoming the infrastructure layer that enables countries to build their own AI ecosystems, rather than being limited to selling GPUs (graphics processing units) alone. This positions NVIDIA as essential to the next phase of AI development, regardless of geopolitical tensions or data sovereignty concerns.