Naver is joining hands with Nvidia to secure leadership in the global artificial intelligence (AI) infrastructure market.

Naver said on 8 it has agreed with Nvidia on a joint project to build a gigawatt (GW)-scale ultra-large global AI factory. The alliance is an integrated partnership spanning all stages of the value chain, including identifying global demand and capital cooperation. Naver will participate as a core partner, jointly responsible for business results and risks.

Lee Hae-jin (이해진), chairman of Naver's board, and Jensen Huang (젠슨 황), Nvidia's chief executive officer, meet on the day at Naver's 1784 headquarters to discuss a detailed roadmap for the joint project and specific cooperation measures for entering global markets. The two companies plan to secure leadership in the AI infrastructure ecosystem across markets including Asia-Pacific as well as Europe and the Middle East.

Naver will use its hyperscale data center, Gak Sejong, as an outpost to absorb global demand. It plans to build GW-scale infrastructure by starting 55MW operations in the first half of 2027, expanding overseas infrastructure to 100MW in the same year and to 200MW in 2028. One GW is about 4 times the maximum capacity of Gak Sejong, enough to accommodate several hundred thousand of the latest graphics processing units (GPUs) simultaneously.

Technical cooperation will also be advanced. Naver will combine its capabilities in building and operating its own GPU clusters and its hyperscale data center know-how with Nvidia's next-generation high-performance infrastructure platform DSX to maximize data center operating efficiency. The companies will also pursue cooperation in spatial intelligence, including building a "Seoul World Model" using Naver's own spatial modeling and Street View data on Nvidia's world foundation model, Cosmos.

Naver has also recently joined the "Nvidia Nemotron Alliance," which includes 12 global AI companies such as Cursor, Mistral AI and Perplexity. Based on this, Naver plans to combine its own data and training know-how to enhance the performance of HyperCLOVA X and accelerate efforts to secure global general-purpose applicability.

Lee said, "Through this alliance, it is encouraging that we can present concrete alternatives so that each region and country around the world can build its own sovereign AI capabilities," adding, "This cooperation is meaningful in that it has created an opportunity for Naver's technological infrastructure competitiveness to take a step forward into the global market."