Nvidia SK Group AI Partnership Expands Data Center Push
Nvidia SK Group AI partnership signals a multi-year push into data centers and memory and could prompt traders to reweight data-center and memory plays.
KEY TAKEAWAYS
- Unveiled a more than $500.0 billion AI initiative to fund data centers and next-generation memory.
- SK Telecom targets a 2 GW Nvidia-powered AI data center using HBM4 with first facility due in 2027.
- A presidential adviser cited roughly $950.0 billion in memory partnerships involving Samsung and SK hynix.
HIGH POTENTIAL TRADES SENT DIRECTLY TO YOUR INBOX
Add your email to receive our free daily newsletter. No spam, unsubscribe anytime.
Nvidia (NVDA) said on July 24, 2026, that its AI partnership with South Korea’s SK Group will fund large-scale AI data-center construction and long-term memory supply agreements, shaping Nvidia’s data-center, personal-AI, and physical-AI roadmaps.
Initiative Scope and Timeline
Nvidia and SK Group unveiled an AI initiative valued at more than $500 billion, aimed at financing large-scale AI data centers and next-generation memory development. The program is a multi-year capital effort to secure compute and memory capacity for large training workloads and new AI applications.
SK Telecom said it will add Nvidia-powered AI cloud capacity built on Nvidia’s DSX AI Factory architecture, with its first AI factory scheduled to come online in 2027. As part of the collaboration, SK Telecom plans a 2-gigawatt AI data center powered by Nvidia’s Vera Rubin processors and SK hynix’s HBM4 high-bandwidth memory.
On June 7, Nvidia and SK hynix launched a multiyear AI-memory partnership that expands their supplier relationship into co-developing next-generation high-bandwidth memory across Nvidia platforms, including Vera Rubin supercomputers, Vera CPUs, RTX Spark PCs, and Jetson Thor robotics. SK hynix is Nvidia’s largest memory supplier.
Memory Supply and National Strategy
A South Korean presidential adviser said on July 25, 2026 KST that Samsung Electronics and SK hynix will pursue memory-chip supply partnerships with U.S. big-tech companies, including Nvidia, worth about $950 billion overall. The adviser linked the initiative to a broader push to expand chip production, data centers, and national AI infrastructure.
Nvidia CEO Jensen Huang met President Lee Jae-myung and outlined expanded cooperation with SK hynix and Samsung on chip and memory design. Nvidia said in a blog post at its AI Summit in Korea that it would work with KAIST on a Korea-specific large-language model. Separately, Nvidia, NAVER, and Brookfield plan to expand NAVER’s AI data-center operations in South Korea. Brookfield is leading a $100 billion AI infrastructure program supported by a $10 billion fund and identifies Nvidia as a partner.
Together, these corporate and government partnerships aim to anchor gigawatt-scale AI factories and localized model development in South Korea, securing long-term access to the compute and memory capacity Nvidia and its partners say is central to the next phase of AI deployment.
