SK Group and NVIDIA announced on 2026-07-24 what NVIDIA's press release describes as "plans for a $500-billion-plus comprehensive partnership to establish AI infrastructure serving the surging demand for global compute" (E1). The stated goal, according to the release, pairs two of SK Group's units with NVIDIA's compute platforms: SK Group Chairman Chey Tae-won said, "By leveraging SK hynix's AI memory and SK Telecom's AI infrastructure capabilities, SK will collaborate with NVIDIA to build a world-class AI factory, helping Korea transcend its role as a leading adopter of AI and become a global hub that drives AI innovation" (E5). NVIDIA founder and CEO Jensen Huang framed the deal in similar terms, saying the companies are "building a new generation of AI factories that will power Korea's next wave of growth" (E6).
Per NVIDIA's announcement, SK Telecom will build a facility described as a "2-gigawatt NVIDIA Vera Rubin DSX AI Factory to serve global compute demand" (E2). NVIDIA's release further specifies that "this cloud will use the NVIDIA DSX platform and deploy NVIDIA Vera Rubin accelerated computing powered by SK hynix HBM4, with the first AI factory planned to come online in 2027" (E3). That timeline places roughly a year between the July 2026 announcement and the first factory's planned activation.
Alongside the compute buildout, NVIDIA and SK hynix (SK海力士) separately "establish long-term partnership to secure and codevelop next-generation AI memory, including HBM," according to the release (E4). This memory arrangement is the supply-side counterpart to the compute deployment described above: the same release ties SK hynix's HBM4 directly to the Vera Rubin accelerated computing that SK Telecom's AI factory will run (E3).
NVIDIA's release identifies SK Group as "South Korea's second-largest conglomerate," with "over 175 affiliate companies and over 100,000 employees worldwide," spanning the artificial intelligence, semiconductors, energy, and life sciences ecosystems (E7). Within that portfolio, the partnership specifically draws on two units named by Chairman Chey: SK hynix for AI memory and SK Telecom for AI infrastructure (E5).
Jensen Huang's quote in the release lays out NVIDIA's stated rationale: "South Korea has all the ingredients to become a global AI powerhouse — world-class networks and data centers, leadership in chip technology and vast industrial scale" (E6). NVIDIA frames this as the basis for building the new AI factories with SK Telecom and SK hynix.
| Metric | Value | Source |
|---|---|---|
| Total partnership scope | $500-billion-plus | NVIDIA press release (E1) |
| SK Telecom AI factory capacity | up to 2 gigawatts | NVIDIA press release (E2) |
| First AI factory online target | 2027 | NVIDIA press release (E3) |
| SK Group affiliate companies | 175+ | NVIDIA press release (E7) |
| SK Group global employees | 100,000+ | NVIDIA press release (E7) |
The figures disclosed in NVIDIA's release describe a partnership built on two linked tracks: a $500-billion-plus umbrella figure (E1) attached to a concrete, dated deliverable — a 2-gigawatt SK Telecom AI factory targeted for 2027 (E2, E3) — and a separate, undated long-term memory-supply agreement between NVIDIA and SK hynix covering HBM4 and future HBM generations (E4). Chairman Chey's quote explicitly binds these two tracks together, crediting SK hynix's memory and SK Telecom's infrastructure as the joint basis for the AI factory (E5), while Huang's quote attributes NVIDIA's choice of partner to South Korea's existing chip-technology leadership and industrial scale (E6) — a scale NVIDIA's own release quantifies via SK Group's 175-plus affiliates and 100,000-plus employees (E7).
NVIDIA's press release describes it as a "$500-billion-plus comprehensive partnership."
Up to 2 gigawatts, built as an NVIDIA Vera Rubin DSX AI Factory, according to NVIDIA.
NVIDIA's release states the first AI factory is planned to come online in 2027.
According to NVIDIA's newsroom, NVIDIA has expanded its NVIDIA Agent Toolkit for engineering by adding NVIDIA PhysicsNeMo and three new CUDA-X libraries — cuISS, cuDSS and cuEST — as agent-ready tools. NVIDIA states partners including Cadence, Siemens, Samsung, Silvaco, Keysight, Synopsys and TSMC have reported speedups ranging from 10x to 50x using the new tools.
According to NVIDIA's official blog, the company is deploying its 88-core Vera CPU across the EDA workflows used to design its next-generation chips, and is working with Cadence and Synopsys on tools that showed up to 1.5x higher performance on selected workloads.
According to CNA and UDN, FSC (金管會) said mortgages reached NT$11.7 trillion (+4.3%), while consumer credit grew faster; overall risk remains manageable.