NAVER, NVIDIA, and Brookfield announced a major expansion of South Korea’s sovereign AI infrastructure, escalating the nation’s computational capacity to support domestic and global AI innovation. The initiative, unveiled during President Lee Jae-myung’s AI Summit delegation in San Francisco, outlines a plan to scale the existing NVIDIA DSX AI factory at NAVER’s GAK Sejong hyperscale facility from 55 megawatts to 200 megawatts by 2028, with a long-term trajectory reaching one gigawatt. The deployment will be financed through a tripartite capital agreement. NVIDIA has committed to a one billion dollar strategic investment in NAVER, contingent upon customary closing conditions and NAVER securing at least nine billion dollars in additional financing. Brookfield has signed a nonbinding term sheet to provide up to nine billion dollars, leveraging its extensive artificial intelligence infrastructure management portfolio. NAVER will cover the remaining capital requirements, accelerating the transition from infrastructure planning to large-scale execution. The expanded facility will deploy advanced NVIDIA accelerated computing architectures, including Blackwell and the upcoming Vera Rubin platforms, optimized through the DSX ecosystem. DSX operating systems will manage multi-tenant operations, lifecycle automation, and system resiliency across the campus, while dedicated throughput software will maximize computational efficiency per megawatt. NAVER will continue operating the hyperscale site as an NVIDIA Cloud Partner, integrating full-stack hardware, software, and facility engineering to minimize token generation costs and accelerate production readiness. Beyond raw compute, the partnership targets sovereign AI development and open-model advancement. NAVER is integrating its HyperCLOVA X models with NVIDIA Nemotron 3 Ultra frameworks and has joined the Nemotron Coalition to contribute to pretraining and reinforcement learning research. The company plans to launch a domestic AI agent platform in the second half of the year, utilizing NVIDIA Agent Toolkit software. Additionally, NAVER is developing a Seoul World Model based on Cosmos foundation models, leveraging proprietary urban spatial data. Corporate leadership emphasized that combining investment capital, operational expertise, and accelerated computing will position South Korea at the forefront of the intelligence era. The expanded infrastructure will provide dedicated production-scale compute for Korean enterprises, government agencies, and international AI developers, reinforcing national competitiveness while establishing a scalable blueprint for sovereign artificial intelligence deployment worldwide.