Nvidia and SK Group Unveil Over $500 Billion AI Data Centers Initiative, Memory Partnership

Technology giant Nvidia and South Korea's leading conglomerate, SK Group, have announced an ambitious artificial intelligence (AI) initiative valued at over $500 billion, set to shape the future of AI infrastructure. This strategic collaboration involves a multi-year partnership encompassing the construction of global AI data centers and the development of next-generation memory technologies. The agreement solidifies Nvidia's leadership in the AI ecosystem while strengthening SK Group's position in the rapidly evolving AI market.

At the core of the initiative is a long-term technology partnership between SK Hynix, SK Group's memory chip manufacturer, and Nvidia. This collaboration aims to develop advanced, next-generation memory chips, particularly High Bandwidth Memory (HBM) technologies, for global AI data centers. SK Hynix will secure memory supply for various Nvidia AI infrastructures, including Vera Rubin AI supercomputers, Vera CPUs, RTX Spark-powered personal computers, and Jetson Thor robotic computing platforms. Furthermore, the two companies will collaborate on applying AI to semiconductor design and manufacturing processes.

SK Telecom, the telecommunications arm of SK Group, also plays a crucial role in this extensive initiative. The company plans to build a gigawatt-scale AI cloud in South Korea using Nvidia's DSX platform. The first of these "AI factories" is expected to become operational in 2027. These data centers will be powered by Nvidia's Vera Rubin chips and SK Hynix's HBM4 memory, aiming to support sovereign, physical, and agentic AI services across South Korea and potentially the broader Asian region.

This monumental collaboration reflects the surging demand in global markets for AI infrastructure and advanced memory chips. The demand for high-bandwidth memory, in particular, has significantly increased with the rapid proliferation of AI technologies, leading to supply shortages in the market. For Nvidia, this partnership guarantees access to critical HBM memory, while further solidifying SK Hynix's position as a key supplier within the AI ecosystem. These developments directly impact company valuations across the semiconductor and technology sectors.

In a broader economic and political context, this development comes during a period of deepening technology collaborations between the U.S. and South Korea, highlighted by Jensen Huang's visits to the country. Nvidia CEO Jensen Huang emphasized the booming semiconductor and industrial sectors in South Korea, stating that the nation possesses the capability to significantly contribute to building global AI infrastructure. This is seen as a strategic move aimed at solidifying regional technology leadership and maximizing benefits from the AI revolution.

Market analysts anticipate that such large-scale collaborations will accelerate the development of AI technologies and further intensify competition. The rivalry among HBM manufacturers like SK Hynix, Samsung, and Micron is expected to heighten with the introduction of next-generation HBM4 memory. In the coming period, AI data center investments are projected to continue their upward trend, and demand for AI chips and memory is expected to remain robust, which could positively reflect on the financial performance of the involved companies.