Naver has set a target margin rate of “20% or higher” for its artificial intelligence (AI) factory business at maturity, drawing attention to whether AI data centers requiring massive capital investment can establish themselves as a genuine profit source. AI data centers require enormous upfront investment in graphics processing units (GPUs), power, and data center construction, and carry a heavy depreciation burden, making it structurally difficult to secure actual profits even as revenue grows rapidly. Against this backdrop, Naver’s target of 20%+ margins is being interpreted as a strong signal of confidence in the profitability of its AI infrastructure business.<\/p>\n
Naver expects revenue from its AI factory venture with Nvidia to begin in earnest in the first half of next year. Last month, Naver announced a capital increase of $1 billion (approximately \u20a91.4 trillion) through a share issuance to Nvidia. The funds will be deployed toward construction of an AI factory targeting 1 gigawatt (GW) of capacity. Nvidia will serve as a strategic investor, while global alternative asset manager Brookfield will invest $9 billion (approximately \u20a912.8 trillion) to become the largest shareholder of a special purpose vehicle (SPV) that holds infrastructure assets including GPUs and data centers. Naver will be responsible for AI factory operations and customer acquisition.<\/p>\n
The profitability targets for the business structure are also specific. Kim Hee-chul, Naver’s CFO, said during an earnings conference call on the 7th of this month: “Revenue, costs, and resulting margins from the computing business are entirely captured by the AI factory operator (Naver’s side).” He explained, “Operating a computing business requires procurement of massive computing assets, but rather than Naver procuring these directly, Brookfield will procure them through the SPV and supply them to Naver.” He added, “The operator’s expected return will vary depending on maturity. Margins may be low initially, but as the business matures, we expect margin rates in the double digits at a minimum, and further, 20% or higher.”<\/p>\n
Comparison with Global AI Infrastructure Players<\/p>\n
Whether Naver’s target is excessively high can be assessed by comparing it with global AI cloud companies. AI-specialized cloud provider Nebius targets margins of 20-30% on an adjusted EBIT (earnings before interest and taxes) basis that reflects GPU depreciation. In its Q1 2025 shareholder letter, Nebius stated: “Assuming a conservative 4-year depreciation schedule over the medium term, our base case anticipates achieving billions of dollars in revenue within a 20-30% adjusted EBIT margin range.” In its Q1 2026 materials, adjusted EBITDA (earnings before interest, taxes, depreciation, and amortization) margin had risen to 45%, which the company described as being “on a trajectory toward 20-30% EBIT margins.”<\/p>\n
EBIT measures operating profitability that reflects depreciation of GPUs, servers, and data centers, while EBITDA shows operating profitability excluding depreciation. AI infrastructure businesses can show a wide gap between these two metrics due to their capital-intensive nature. Massive upfront investment in GPUs, servers, and data centers generates substantial related depreciation expenses. For this reason, it is difficult to simply evaluate the profitability of AI infrastructure businesses on an EBITDA basis alone.<\/p>\n
The case of AI-specialized cloud provider CoreWeave illustrates this well. CoreWeave posted adjusted EBITDA of $1.51 billion (approximately \u20a92.1 trillion) in Q2 2026, but adjusted operating income was only $128 million (approximately \u20a9180 billion). Based on revenue of $2.575 billion (approximately \u20a93.7 trillion), adjusted EBITDA margin was approximately 59%, but adjusted operating margin reflecting depreciation of GPUs and data centers was only about 5%. This demonstrates that even when AI computing demand surges and revenue and EBITDA grow rapidly, actual operating profitability that accounts for depreciation of invested assets does not necessarily improve at the same pace.<\/p>\n
If the 20% Naver is referring to is on an EBITDA basis, it would not be considered a particularly high target compared with global AI infrastructure operators. However, if it refers to operating margin that reflects depreciation of GPUs and data centers, it could be viewed as a considerably aggressive target.<\/p>\n
Of course, there are limitations to simply comparing Naver’s margin target with global players. Naver has not specified whether the 20%+ margin rate is based on EBITDA, adjusted EBIT, operating margin, or some other metric. The business structures also differ. Nebius and CoreWeave are AI infrastructure specialists that provide AI cloud services based on GPUs and data centers. Naver, by contrast, operates not only its own AI models and cloud services but also existing platform businesses including search, advertising, and commerce. The ability to leverage computing capacity secured from the AI factory for its own services and enterprise AI offerings also creates differences in the revenue model.<\/p>\n
Securities Analysts’ Outlook and Key Variables<\/p>\n
Securities analysts are focusing on the potential for the AI factory to become Naver’s new growth engine. Ahn Jae-min, an analyst at NH Investment & Securities, said: “Once the AI data center business ramps up in earnest and related revenue begins to materialize from 2027, the value of this new business is expected to be recognized. We estimate Naver’s AI factory revenue will grow from \u20a9560 billion (approximately $394.7 million) in 2027 to \u20a96.4 trillion (approximately $4.5 billion) in 2030, with operating profit expected to grow from \u20a931.7 billion (approximately $22.3 million) in 2027 to \u20a9720.5 billion (approximately $507.9 million) in 2030.” Calculating NH Investment & Securities’ projections as an operating margin yields a figure in the low 10% range.<\/p>\n
Ultimately, the actual profitability of the AI factory will depend on how efficiently Naver manages utilization rates, GPU costs, power costs, depreciation, and customer acquisition costs. For Naver to actually achieve its stated 20%+ margin target, the key variables will be how quickly it absorbs the initial investment burden and how high it can push GPU and data center utilization rates.<\/p>\n
Lee Kyung-jun, a professor in the Department of Big Data Applications at Kyung Hee University, said: “Naver will operate the AI factory competitively, but how well it secures customers (demand) will determine the AI factory’s future profitability.”<\/p>\n","protected":false},"excerpt":{"rendered":"Naver has set a target margin rate of “20% or higher” for its artificial intelligence (AI) factory business…\n","protected":false},"author":2,"featured_media":118827,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[623],"tags":[38995,11933,48359,26985,48478,61496,657,49803,986,2295],"class_list":["post-118826","post","type-post","status-publish","format-standard","has-post-thumbnail","category-naver","tag-ahn-jae-min","tag-ai-factory","tag-brookfield","tag-coreweave","tag-kim-hee-chul","tag-lee-kyung-jun","tag-naver","tag-nebius","tag-nh-investment-securities","tag-nvidia"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/posts\/118826","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/comments?post=118826"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/posts\/118826\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/media\/118827"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/media?parent=118826"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/categories?post=118826"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/korea\/wp-json\/wp\/v2\/tags?post=118826"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}