Skip to main content

South Korea says Nvidia reaffirmed its 260,000-GPU supply plan

14 NOVEMBER 2026·2 MIN READ·4 SOURCES

South Korea’s science minister said Nvidia confirmed it would continue its planned supply of up to 260,000 GPUs to the country, despite Jensen Huang stepping down.

Key takeaways · 4

  • 01

    The minister said Nvidia confirmed a plan to supply up to 260,000 GPUs to South Korea.

  • 02

    The original plan involved the government, Samsung Electronics, SK Group, Hyundai Motor Group and Naver Cloud.

  • 03

    The government planned to procure about 35,000 GPUs directly; Bae said about 21,000 had been secured.

  • 04

    MLQ.ai reported that about 14,000 more directly procured GPUs were scheduled for delivery by early 2027.

Minister reports Nvidia reaffirmation

South Korea’s science minister said Nvidia confirmed its commitment to deploy up to 260,000 GPUs in the country, and that its planned supply would proceed despite Huang’s resignation.[1] Bae said in a social media post titled “Jensen Huang's Resignation, but the Cooperation Continues” that Nvidia reaffirmed the supply and cooperation plans.[2] The science ministry held a meeting with Nvidia to review pending issues, according to Bae.[1] MLQ.ai reported that the ministry’s second vice minister chaired a review meeting with Nvidia on Friday morning.[2]

Plan spans government and industry

Nvidia announced the supply plan during the APEC meetings in Gyeongju in October 2025.[2] The original plan covered the South Korean government and Samsung Electronics, SK Group, Hyundai Motor Group and Naver Cloud.[1] Nvidia described the planned infrastructure as supporting AI-enabled economic growth and innovation across Korean industries, including automotive, manufacturing and telecommunications.[3] The original announcement said the GPUs would support AI infrastructure across sovereign clouds and AI factories.[3]

Procurement figures and intended uses

Under the original plan, the South Korean government aimed to deploy up to 50,000 of Nvidia’s latest GPUs for sovereign AI development.[3] MLQ.ai reported allocations of about 50,000 Blackwell GPUs each for Samsung, SK Group and Hyundai Motor Group, and 60,000 for Naver Cloud.[2] The government planned to procure about 35,000 GPUs directly, and Bae said about 21,000 had been secured.[2] About 14,000 more directly procured GPUs were scheduled for delivery by early 2027.[2] The reported quantities are measured in performance equivalent to Nvidia’s B200.[2]

Confirmation remains attributed to officials

The reported commitments are not the same as a new public statement from Nvidia: MLQ.ai said the company had not issued its own statement on the Korea commitments after Huang stepped down.[2] The government said the GPUs would support AI research by companies, universities and research institutes, as well as policy projects including Korea’s sovereign AI foundation model.[2] The reported cooperation agenda also includes linking Korean-made AI semiconductors with Nvidia’s platform using NVLink, building AI computing infrastructure, joint physical AI research and talent development, and pursuing a GTC event in South Korea.[4]

For teams planning AI infrastructure, the reported pledge provides a scale and a set of intended recipients to consider, but the confirmation in this account comes from South Korean officials. Professionals should distinguish the broader 260,000-GPU commitment from the government’s separate direct-procurement figures when assessing the plan.

Why it matters
For teams

AI training for your team

kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.

How this developed

  1. 14 November 2026

    South Korea says Nvidia reaffirmed its 260,000-GPU supply plan

Sources

AI fluency, one session a day, built for your work.