The partnership covers AI data centers, next-gen memory chips, and a 2-gigawatt facility set to go live in 2027.
Nvidia just signed one of the largest tech partnership agreements in recent memory. On July 25, 2026, the chipmaker and South Korea’s SK Group inked letters of intent for a collaboration valued at over $500 billion, covering AI data centers, memory chip supply, and next-generation computing infrastructure.
What the deal actually covers #
The partnership has three main pillars. First, SK Telecom will build a 2-gigawatt AI data center powered by Nvidia’s Vera Rubin platform, with the first phase targeted for operation in 2027. Second, Nvidia and SK hynix will deepen a long-term collaboration to co-develop HBM4, the next generation of high-bandwidth memory that AI training workloads depend on. Third, the broader SK Group relationship is designed to address memory supply shortages that have quietly become one of the biggest bottlenecks in scaling AI systems globally.
The Nvidia-SK hynix technology partnership was first announced on June 7, 2026, focused on aligning memory roadmaps with Nvidia’s AI hardware plans. The July deal formalizes that relationship into something far bigger and more capital-intensive.
In May 2026, Nvidia signed a separate agreement with IREN to deploy up to 5 gigawatts of Nvidia DSX infrastructure globally.
Why memory supply is the real story here #
SK hynix is one of only a handful of companies on earth capable of producing HBM at scale, alongside Samsung and Micron. Locking in a long-term co-development agreement with SK hynix is Nvidia effectively buying insurance against future supply crunches.
The 2-gigawatt data center figure deserves attention on its own. Most hyperscale data centers operate in the hundreds of megawatts range. A 2-gigawatt facility would be among the largest AI compute installations anywhere in the world, and it is being built around Nvidia’s own hardware platform.
What this means for crypto and AI token markets #
Projects like Bittensor (TAO), Render (RNDR), and Fetch.ai (FET) sit in the category most likely to see renewed interest following announcements like this one. These networks are positioning themselves as decentralized alternatives to or complements of centralized AI infrastructure.
Investors watching this space should weigh a few risks carefully. AI token valuations are highly sensitive to broader crypto market conditions, and a sentiment shift can unwind AI-narrative gains quickly. The tokens themselves represent early-stage networks, not mature infrastructure businesses.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our