SourceBloomberg Technology•Stalecollected in 8m
Jensen Huang's High-Profile South Korea Visit
💡See how Nvidia's CEO is strengthening ties with key Asian tech hubs to maintain market dominance.
⚡ 30-Second TL;DR
What Changed
Jensen Huang met with esports legend Faker
Why It Matters
The visit reinforces Nvidia's strategic importance to the South Korean semiconductor and AI ecosystem. It signals strong executive-level alignment between Nvidia and key regional partners.
What To Do Next
Monitor Nvidia's partnership announcements with Korean firms for potential supply chain shifts in AI hardware.
Who should care:Founders & Product Leaders
Key Points
- •Jensen Huang met with esports legend Faker
- •Engaged in networking with South Korean tech titans
- •Demonstrated significant public and industry influence in Seoul
🧠 Deep Insight
Background and context from public sources — not the original article. 29 sources cited.
🔑 Enhanced Key Takeaways
- •Jensen Huang's visit solidified multi-billion dollar AI infrastructure and memory co-development deals with major South Korean conglomerates, including SK Group, LG, Naver, and Doosan, focusing on gigawatt-scale AI factories, physical AI, and advanced memory components.
- •Nvidia unveiled RTX Spark, a new superchip combining a 20-core Grace ARM CPU and Blackwell GPU with 1 petaFLOP of FP4 AI performance, designed for Windows laptops and compact desktops, with a planned launch in Fall 2026.
- •The visit underscored South Korea's strategic importance as a global hub for AI and memory chip production, with the country's government already investing in sovereign AI infrastructure with over 50,000 NVIDIA GPUs.
- •Nvidia's venture arm, NVentures, made its first investment in a Korean startup, Point2 Technology, in April 2026, backing its "e-Tube" data transmission technology for AI data centers.
📊 Competitor Analysis▸ Show
| Feature/Category | Nvidia (RTX Spark, GPUs) | FuriosaAI (RNGD) | Rebellions (Rebel Quad) | AMD (MI300, Radeon GPUs) | Intel (CPUs, GPUs) | SambaNova Systems (SN50) |
|---|---|---|---|---|---|---|
| Primary Focus | AI Training & Inference, Gaming, Physical AI, Data Centers | AI Inference (energy efficiency) | AI Inference (consistent performance, mixed-precision) | AI Training & Inference, Gaming | CPUs, AI Training & Inference, Gaming | High-volume Generative AI, Agentic AI |
| Architecture | GPU (Blackwell), Grace ARM CPU (RTX Spark) | Neural Processor (ASIC-like) | Chiplet-based SoC (UCIe advanced channels) | GPU | CPU, GPU | Reconfigurable Data Unit (RDU) |
| Key Differentiator | Full-stack accelerated computing, broad ecosystem, market dominance | Optimized for inference, claims 2.25x inference performance vs. competing GPUs with LG's EXAONE AI model | Inference-first hardware, H200 compatible, UCIe advanced channels, tight memory integration | Competing for AI training market share, general-purpose GPUs | Own foundry, strong CPU market, emerging GPU presence | Claims 5x faster max speed, 3x lower TCO for agentic AI vs. competitive chips |
| South Korea Market Share (Gaming GPU) | ~76% (as of June 2025) | Emerging, tested by LG | Early deployments in South Korea's largest telco | ~21% (as of June 2025) | ~2-3% (as of June 2025) | Not specified |
| Availability | RTX Spark: Fall 2026 | Mass production | Mass production | Available | Available | SN50 unveiled Feb 2026 |
🛠️ Technical Deep Dive
- NVIDIA RTX Spark: A superchip designed for Windows laptops and compact desktops. It integrates a 20-core Grace ARM CPU and a Blackwell GPU on a single TSMC 3nm package. It features up to 128 GB of shared LPDDR5X memory and delivers 1 petaFLOP of FP4 AI performance.
- FuriosaAI RNGD (Renegade): A neural processor specifically designed for AI inference workloads. It focuses on matrix computations and parallel operations, aiming for high energy efficiency.
- Rebellions Rebel Quad: A datacenter inference chip featuring a chiplet-based SoC design that combines four chiplets into a coordinated unit. It utilizes three UCIe advanced channels per chiplet (total of 12) for high-speed die-to-die communication, with one channel for HBM and two for adjacent chiplets in a 2x2 grid. It is positioned as H200 compatible and is the world's first UCIe advanced-based AI accelerator.
🔮 Future ImplicationsAI analysis grounded in cited sources
Nvidia will further entrench its dominance in the global AI infrastructure market.
The extensive partnerships with South Korean tech giants for gigawatt-scale AI factories and memory co-development secure critical supply chains and expand Nvidia's ecosystem, particularly in Asia.
South Korea will emerge as a leading hub for 'physical AI' and AI-driven manufacturing.
Nvidia's collaborations with LG, Hyundai, and Doosan specifically target robotics, autonomous driving, and industrial automation, leveraging Korea's manufacturing prowess and integrating AI systems into real-world applications.
Competition in the AI inference chip market will intensify, particularly from South Korean startups.
Companies like FuriosaAI and Rebellions are actively developing specialized AI inference chips, aiming to challenge Nvidia's market share with claims of better efficiency and performance for specific workloads, offering alternatives to Nvidia's general-purpose GPUs.
⏳ Timeline
2025-10
Jensen Huang's previous visit to Korea, meeting Samsung and Hyundai executives; Korean government announced $3 billion investment in physical AI cluster with NVIDIA GPUs.
2026-04
NVentures, Nvidia's venture arm, made its first investment in a Korean startup, Point2 Technology, for its e-Tube data transmission technology.
2026-06
Jensen Huang visited South Korea, unveiling RTX Spark, meeting with esports legend Faker, and engaging with major tech leaders to announce significant AI infrastructure, memory, and robotics partnerships.
📎 Sources (29)
Factual claims are grounded in the sources below. Forward-looking analysis is AI-generated interpretation.
- nvidia.com
- intellectia.ai
- nvidia.com
- morningstar.com
- turkiyetoday.com
- almayadeen.net
- investing.com
- thehindu.com
- chosun.com
- techtimes.com
- businessinsider.com
- nvidia.com
- gurufocus.com
- sedaily.com
- joins.com
- pcmag.com
- youtube.com
- aimultiple.com
- reddit.com
- voi.id
- joins.com
- chosun.com
- joins.com
- tweaktown.com
- chosun.com
- koreaherald.com
- youtube.com
- koreaherald.com
- investing.com
📰
Weekly AI Recap
Read this week's curated digest of top AI events →
👉Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: Bloomberg Technology ↗
This is a summary, not the original. Read the source, or get the weekly briefing.
The weekly digest
One email a week. Unsubscribe anytime.