Naver has secured a computing services supply contract worth about 2 trillion won for its AI Factory business. It plans to start commercial services in the first half of next year, generating revenue, and to step up its computing infrastructure business using next-generation graphics processing units (GPUs).

Naver Chief Executive Choi Soo-yeon (최수연) said on Nov. 12 during a conference call on the company’s third-quarter 2026 earnings that it signed a computing services supply contract worth about 2 trillion won with a U.S. AI inference company, including fourth-generation GPUs. She said the contract completes customer acquisition for an initial 2027 supply volume.

Choi said, "This contract is a meaningful deal that confirms actual market demand and early business viability for the AI Factory business." She added, "Revenue is also expected to occur sequentially along with the launch of commercial services in the first half of next year."

Naver plans to pursue mid- to long-term contracts with large customers while expanding its customer base to global AI companies with high growth potential. It is discussing additional supply contracts with multiple companies, including large AI model developers, and expects contract expansions as existing customers’ computing demand increases.

It is also working to secure infrastructure to meet customer demand. Choi said data center capacity secured through contracts so far has exceeded the company’s previously disclosed 2028 target volume.

Naver said it has preemptively secured most liquid-cooled data centers available for use in South Korea. Because next-generation GPUs require high power supply capacity and cooling facilities, it expects securing related infrastructure to work in its favor in attracting global customers.

It also plans to expand GPU procurement. Naver is in talks with Nvidia over what it described as the largest server order contract in Asia. It aims to lower procurement costs and secure supply capacity for 'Vera Rubin,' a next-generation GPU platform.

It has also put in place an investment and operating base. Naver signed a computing infrastructure investment agreement with Brookfield and completed participation in Nvidia’s paid-in capital increase. It plans to use external capital and technology cooperation to reduce the burden of initial investment, while applying Naver Cloud’s experience in building and operating GPU clusters to the business.

Naver plans to shorten the time from introducing GPU equipment to commercial service by combining Nvidia’s standard architecture with its own operating capabilities. It expects this will help it meet supply schedules required by customers and bring forward the timing of revenue generation.

Choi said, "Going forward, based on a favorable market environment and the fourth-generation GPU supply capacity we have secured in advance, we will continue to secure customers and deliver meaningful revenue and profit results in the AI Factory business through high utilization and profitability."