Lee Seeks Tech Support for Korea AI Projects

At the summit, Lee unveiled the San Francisco AI Declaration, outlining South Korea's ambitions for technological cooperation with the United States, Reuters reported. In remarks quoted by Reuters, Lee said he wanted to present a vision for AI's future and propose global cooperation to make it happen. SBS reported that Lee described South Korea as seeking to become a global AI production base, an innovation foundation and a platform for AI cooperation.

The agenda connects AI model development and deployment to the supply of accelerators, memory, data-center capacity and related power infrastructure. Reuters framed the diplomatic push against rapidly growing demand for chips and compute, which has increased the importance of South Korea's semiconductor and memory suppliers in the AI supply chain.

For AI practitioners, national AI programs of this scale are consequential when they result in accessible compute, domestic cloud capacity, research collaborations and production-grade deployment partnerships. Comparable public-private infrastructure initiatives often create opportunities for model training, inference optimization and sovereign-data workloads, while their practical value depends on procurement details, capacity allocation and operating economics that were not specified in the reported accounts.

At the summit, Huang said Nvidia and SK Group would announce a business partnership valued at $500 billion, SBS reported. Reuters likewise reported that Nvidia and SK unveiled a $500 billion-plus initiative spanning large-scale AI data centers and next-generation technology, although the retrieved Reuters text did not provide the full terms.

Korea JoongAng Daily reported that Huang referenced potential cooperation with Korean companies involving chip and memory design with SK hynix and Samsung Electronics, autonomous Genesis vehicles and robotics with Hyundai Motor Group, and power-generation systems with LG. Those reported areas are broad, and the available coverage does not establish product specifications, deployment schedules or binding commercial commitments for each project.

Nvidia also announced a joint AI research laboratory with the Korea Advanced Institute of Science and Technology (KAIST) at KAIST's Kim Jaechul Graduate School of AI in Seoul, according to Nvidia's corporate blog. Nvidia described it as its first joint AI lab with a Korean university and said the work would focus on agentic AI research.

The combination of an industry summit, proposed infrastructure investment and an academic lab places the event across multiple layers of the AI stack. Publicly reported collaborations can improve a country's access to hardware and research expertise, but practitioners will need subsequent disclosures on GPU volumes, data-center buildouts, cloud availability, model governance and research outputs to assess their operational significance.