Nvidia and SK Group announce a $500 billion joint investment plan for AI infrastructure development in South Korea, representing the largest single-tranche commitment to the sector.

Nvidia and South Korea's SK Group announced a major partnership on July 24 valued at more than $500 billion, spanning large-scale AI data centers and next-generation memory production — one of the largest single AI infrastructure commitments made to date.

At the center of the deal is a long-term agreement with SK Hynix, the world's largest supplier of high-bandwidth memory, to secure next-generation memory supply for Nvidia and jointly develop HBM for AI training, agents, and physical AI applications. As part of the initiative, SK Telecom plans to build a 2-gigawatt AI data center powered by Nvidia's Vera Rubin chips and SK Hynix's HBM4 memory, with the first facility due online in 2027.

In parallel, Nvidia is investing $1 billion directly into Naver, South Korea's dominant search and internet company, to help finance an AI data center under construction. The investment will enable Naver to more than triple the facility's planned size, from 55 megawatts to 200 megawatts. U.S. private equity firm Brookfield has agreed to a nonbinding term sheet to fund up to $9 billion of the project.

The announcements coincided with a Silicon Valley visit by South Korean President Lee Jae Myung, during which Nvidia CEO Jensen Huang described the country as entering a 'golden age,' citing its semiconductor manufacturing base as a natural fit for AI infrastructure expansion.

The $500 billion figure encompasses both Nvidia's planned purchases of memory from SK Hynix and SK Group's own purchases of Nvidia supercomputing hardware, representing a two-way trade relationship rather than a one-time investment.

This deal builds on agreements Nvidia, SK Telecom, and SK Hynix struck in June 2026, which laid the groundwork for the multi-gigawatt AI infrastructure buildout now underway across South Korea.

High-bandwidth memory has become one of the tightest constraints in the global AI supply chain. By aligning SK Hynix's production roadmap directly with Nvidia's chip release schedule, the partnership is designed to prevent memory shortages from bottlenecking future GPU generations.