NVIDIA's Jensen Huang on Building the Dynamo of the Intelligence Age

Watch on YouTube ↗  |  June 10, 2026 at 12:00  |  41:21  |  Sequoia Capital
Speakers
Jensen Huang — CEO, NVIDIA

Summary

Jensen Huang, NVIDIA's founder and CEO, argues that computing is shifting from retrieval-based data centers to generative AI factories that turn electrons into tokens/intelligence. He lays out a five-layer investment framework—energy, chips, infrastructure, models, and applications—and says the AI buildout could grow toward a $20 trillion annual ecosystem. He also rejects AI job-apocalypse narratives, arguing AI elevates tasks and demand, and urges broad engagement with the technology.

  • Computing is moving from retrieval to real-time generation, requiring massive AI factories.
  • NVIDIA's AI racks are described as expensive, high-volume machines producing valuable tokens.
  • Huang presents a five-layer AI investment stack: energy, chips, infrastructure, models, applications.
  • Energy is called the greatest multi-generational opportunity, with nuclear, wind, solar, hydrogen and named equipment suppliers highlighted.
  • Infrastructure layers like land, power, data centers, and financing are described as scarce today.
  • Huang sees physical AI, robotics, and self-driving as an $80 trillion frontier beyond language models.
  • He rejects job-loss fears and argues AI adoption will elevate workers and close the technology divide.
Ideas
Jensen Huang CEO, NVIDIA 3:41
Agentic AI does paid useful work.
AI has evolved from generative chatbots to agentic systems that reason, use tools, and do useful work. Because it can do work, businesses pay for it by the hour, making it the fastest-growing software business in history.
AI software HIGH
Jensen Huang CEO, NVIDIA 14:27
NVIDIA builds AI factory infrastructure.
Computing is shifting from retrieval to real-time generation, requiring massive AI factories that turn electrons into tokens/intelligence. NVIDIA builds these machines: a rack has 72 chips, weighs two tons, costs about $4 million, and a 1 GW AI factory costs about $50 billion but can generate $300-400 billion in intelligence, implying extremely fast ROI and a massive infrastructure buildout.
NVDA HIGH
Jensen Huang CEO, NVIDIA 20:01
Energy layer is massive AI opportunity.
The bottom layer of the AI five-layer cake is energy. Jensen says this is the single greatest opportunity in several generations for energy industries, and any energy source—nuclear, wind, solar, hydrogen, sustainable—will get funded because AI needs enormous power. He cites Siemens, Mitsubishi, and GE Vernova as beneficiaries already doing well.
XLE ICLN NUCLEAR 7011.T FAN SOLAR GEV SIEMENS HIGH
Jensen Huang CEO, NVIDIA 20:58
AI chip and networking layer grows.
The second layer of the AI industrial cake is chips, computers, networking, switches, and silicon photonics. Jensen places this hardware layer inside the investment framework for the AI buildout, which is attracting roughly $1 trillion this year across the five layers.
SMH HIGH
Jensen Huang CEO, NVIDIA 21:08
AI infrastructure is scarce and investable.
The third layer is infrastructure: land, power, shell, financing, and data center operations. Jensen says every one of these is in scarce supply today, making infrastructure a core bottleneck and investment layer for AI.
AIQ SRVR HIGH
Jensen Huang CEO, NVIDIA 21:25
Model layer is huge opportunity.
The model layer sits atop cloud infrastructure and includes OpenAI and Anthropic. Jensen calls it the greatest opportunity in recent human history and says market-driven investment is flowing naturally into the AI ecosystem.
OPENAI AI-SECTOR ANTHROPIC HIGH
Jensen Huang CEO, NVIDIA 24:43
Physical AI is next $80T frontier.
AI can learn any structured domain—proteins, genes, cells, human anatomy, physics, climate, 3D worlds, robotics, and self-driving cars. Jensen says the physical part of the economy is about $80 trillion and is the most important frontier, beyond just language-model chatbots.
Physical AI ROBO DRIV HIGH
Jensen Huang CEO, NVIDIA 24:53
AI application startups attract record capital.
The top layer is applications that enhance the human condition. Jensen says $100 billion of venture capital went into this layer last year—the largest VC year in history—targeting areas such as financial services, legal, accounting, transportation, and logistics.
AI applications HIGH
Up Next
1:13:01
Adam May on $ABVX's blowout data and subsequent stock crash
Yet Another Value Podcast · Jun 10
37:55
Even if you buy good stocks, you'll fail if you invest with borrowed money | Cha Young-joo, Director of Wise Economic Research Institute [Focus Today's Stocks]
3PRO TV (삼프로TV) · Jun 10
49:11
Tech Selloff Deepens Ahead of CPI; US, Iran Exchange Strikes | Bloomberg Brief 6/10/2026
Bloomberg Markets · Jun 10
4:41:21
[26.06.10 Afternoon Broadcast Full View] Historic Rebound Plunges in One Day... 'Volatile' Domestic Stock Market, Margin Calls Pour In [Closing Bell Live]
3PRO TV (삼프로TV) · Jun 10
31:43
오후 2시부터 외인 선물 2조 매수.. 옵션 만기일 상승에 베팅 했나? "근원 CPI가 중요한 이유" / 외인 단타로 휘청이는 한국증시 ㅣ 이권희 대표
815 Money Talk (815머니톡) · Jun 10

This Sequoia Capital video, published June 10, 2026, features Jensen Huang discussing AI software, NVDA, XLE, ICLN, NUCLEAR, 7011.T, FAN, SOLAR, GEV, SIEMENS, SMH, AIQ, SRVR, OPENAI, AI-SECTOR, ANTHROPIC, Physical AI, ROBO, DRIV, AI applications. 8 trade ideas extracted by AI with direction and confidence scoring.

Speakers: Jensen Huang  · Tickers: AI software, NVDA, XLE, ICLN, NUCLEAR, 7011.T, FAN, SOLAR, GEV, SIEMENS, SMH, AIQ, SRVR, OPENAI, AI-SECTOR, ANTHROPIC, Physical AI, ROBO, DRIV, AI applications