Naver will move to break into the global artificial intelligence (AI) data center market on the back of Nvidia's $1 billion equity investment and $9 billion in infrastructure financing through global alternative asset manager Brookfield. It plans to expand its AI business—previously centered on in-house services—into an "AI factory" that supplies GPUs and data centers externally, with the aim of generating revenue starting in 2027.
Naver said it will carry out a third-party allotment paid-in capital increase of about 1.4809 trillion won to Nvidia. Nvidia will acquire 7,241,564 new shares to secure 4.5% equity in Naver. The payment date is Oct. 30.
This is Naver's first attempt at a third-party allotment paid-in capital increase in 22 years and the first since its transfer listing to KOSPI in 2008. The investment is a follow-up measure that gives shape to the "gigawatt-class global AI factory construction agreement" the two companies signed in June. It is meaningful in that the companies have built a strategic alliance that goes beyond a simple GPU supply relationship to share interests in the AI infrastructure business.
Naver will also retire treasury shares in parallel to minimize the dilution of existing shareholders' equity resulting from the new share issuance. The company decided to cancel 4.9 million treasury shares worth about 1 trillion won on Aug. 3.
Brookfield, a global alternative investment firm, will participate in securing data centers and GPUs. Naver selected Brookfield as the exclusive preferred bidder to pursue a $9 billion infrastructure financing program. Through the partnership with Brookfield, it plans to preemptively secure the GPUs, data centers, and power infrastructure needed for a hundreds-of-megawatts (MW) scale AI factory.
Naver Cloud is also discussing an "AI Compute Partnership" with Nvidia. Beyond technical collaboration, it plans to establish a joint business structure that combines capital and customer demand to bring Nvidia in as a key partner for the global AI factory business.
Naver touts as a strength its ability to directly handle the entire process—from data center design and operations, to GPU cluster management, AI platforms, and its own models and services. It has secured GPU installation space at about 20 sites nationwide and aims to shorten the time from equipment procurement to service launch to within a month through a standardized server system.
The business will start in the first half of 2027 with a capacity of 55 MW. Naver plans to expand operations to 200 MW in 2028 and then scale to the 1 GW level to target sovereign AI demand in the Asia-Pacific, Europe, and the Middle East.
Choi Soo-yeon, Naver's CEO, said, "With this collaboration, we have laid the groundwork to expand Naver's technology and data center operations capabilities into the global market," and added, "We will also accelerate customer acquisition to leap forward as a global AI infrastructure company."