SI Agent News
ConfirmedChipsBusiness

NVIDIA announces investments in Korean AI startups

NVIDIA is funding several Korean companies to bolster the local AI ecosystem across infrastructure, models, and robotics.

1,000Korea-based members of NVIDIA Inception

The short version

Why it matters

This signals a strategic push by NVIDIA to deepen its footprint in the Korean tech market. It provides Korean startups with the capital and technical expertise needed to scale their AI products globally.

The MSIT-NVIDIA Partnership

NVIDIA recently hosted the MSIT-NVIDIA AI Startup Partnership Forum at COEX in Korea. The event was organized in collaboration with Korea’s Ministry of Science and ICT to foster growth within the local technology sector. It served as a primary venue for high-level discussions between government officials and industry leaders.

Ryu Je-myung, the vice minister of science and ICT, joined NVIDIA executives to engage with the local ecosystem. The forum brought together over 20 Korean startups that are developing technologies in several key areas. These include foundation models, physical AI, AI infrastructure, and AI semiconductors.

The initiative aims to strengthen the relationship between the Korean government and one of the world's leading AI hardware providers. By facilitating these meetings, the partnership seeks to identify and support the most promising innovations coming out of the region. This cooperation highlights Korea's growing role as a hub for specialized AI development.

Infrastructure and Connectivity

One of the primary investment announcements involved Point2, a company focused on AI interconnects. These components are essential for moving data between different chips and racks within large-scale AI factories. Point2 offers a scalable, radio frequency-based solution for these tasks.

This technology provides a different approach compared to traditional copper or optical interconnects. As AI clusters scale up, they require high levels of reach, performance, and efficiency to function correctly. Point2's solution is designed to meet these specific demands for next-generation infrastructure.

By investing in this space, NVIDIA is addressing the physical limitations of current data centers. Efficient data movement is a bottleneck for training massive models. Supporting a company that offers an innovative alternative to standard wiring helps ensure that the underlying hardware can keep pace with software demands.

Models and Robotics

NVIDIA also announced plans to invest in Motif Technologies, a Seoul-based AI lab. This company focuses on building foundation models designed to power AI agents capable of performing complex tasks. One of their notable releases is Motif-3, which is a large language model provided with open weights.

The investment list also includes RLWRLD, a physical AI startup with a presence in San Francisco, Seoul, and Tokyo. This company develops robotics foundation models that allow humanoid and industrial robots to handle intricate real-world movements. These tasks require five-finger dexterity, such as threading nuts onto bolts or kitting parts from moving conveyors.

These two companies represent the software and physical application layers of the AI stack. While Motif Technologies works on the cognitive side of AI agents, RLWRLD focuses on the physical execution of tasks. Both are critical for moving AI out of the digital realm and into practical, physical environments.

Enterprise AI and Tools

Upstage is another Korean enterprise AI company that received an investment announcement from NVIDIA. Upstage specializes in creating the Solar language models and various document AI tools. These products are designed to help businesses automate their internal operations and manage data more effectively.

The company's tools are specifically built to help organizations extract information from complex documents. By streamlining these workflows, Upstage helps businesses build custom AI applications for their everyday operations. This allows companies to turn unstructured data into usable insights without manual intervention.

The diversity of these investments shows that NVIDIA is looking at the entire lifecycle of AI. From the hardware interconnects provided by Point2 to the enterprise software of Upstage, the goal is to support every layer of the technology. This broad approach ensures that Korean startups can find a place in the global AI economy.

The Inception Program

These new investments are part of a larger, ongoing effort by NVIDIA to support the Korean startup community. The company already works with more than 1,000 Korea-based members through its Inception program. This program is designed to help early-stage startups grow by providing them with specific resources.

Participants in the Inception program receive several benefits to help them scale their businesses. These include access to preferred pricing on hardware, as well as technical expertise from NVIDIA's engineers. The program also provides opportunities for startups to be introduced to potential investors.

By combining these established resources with new direct investments, NVIDIA is deepening its commitment to the region. This strategy helps create a pipeline of innovation where local startups can leverage global expertise. It ensures that Korean developers have the tools they need to compete on an international scale.

What to watch

Try it

Look up the 'NVIDIA Inception' program to see what types of resources are available for early-stage AI startups.

Sources and links

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

Listen to the AI Wire headlines

Listen to the AI Wire

Listen to our daily AI shows every morning: on your commute, in the shower, wherever you are

Listen to the daily shows

First reported by blogs.nvidia.com · Read the original. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

More AI news