Initiative overview
South Korea's Ministry of Science and ICT (MSIT) is investing approximately KRW 1.46 trillion (around USD 1.1 billion) to procure 13,000 high-performance GPUs as part of a national effort to expand domestic AI computing infrastructure and reduce reliance on overseas resources. The initiative is part of South Korea's broader strategy to accelerate AI innovation and strengthen the competitiveness of local technology companies. Following the allocation of a supplementary budget in May 2025, the ministry conducted a month-long selection process involving public bidding, proposal assessments, and on-site inspections, before selecting Naver Cloud, NHN Cloud, and Kakao as project operators.
NHN Cloud manages the largest allocation of 7,656 B200 units with water-cooling systems, of which 6,120 are government-owned and organised into 510-node and 255-node clusters. Naver Cloud oversees 3,056 H200 units, with 2,296 allocated to the government in clusters aimed at universities and research institutions, with full-scale services expected by end of 2025. Kakao deploys 2,424 B200 GPUs, with 2,040 government-operated in a single 255-node cluster, with partial beta services planned before year-end. Over 10,000 GPUs in total are allocated for government-led projects, with remaining units supporting the operational needs of the three providers.
To ensure equitable access, the ministry and operators will establish a unified online GPU support platform through which universities, research institutions, and industry players can apply, with allocations determined by proposal evaluations and project priorities. Idle resources will be dynamically reassigned to maximise utilisation. Once fully operational, the infrastructure will underpin the government's AI Foundation Model Project and other national initiatives. Minister Bae Kyung-hoon described the project as a first step toward building "an AI superhighway" and establishing South Korea as a global AI leader.
