Podcast Summary: "Winning the AI Race Part 3: Jensen Huang, Lisa Su, James Litinsky, Chase Lochmiller"

Release Date: July 23, 2025
Podcast: All-In with Chamath, Jason, Sacks & Friedberg
Hosts: Chamath Palihapitiya, Jason Calacanis, David Sacks & David Friedberg
Guests: Jim Latinsky (CEO of MP Materials), Steve (TSMC representative), Chase Lockmiller (CEO of Crusoe), Jensen Huang (NVIDIA), Lisa Su (AMD)

Introduction

In the third installment of the "Winning the AI Race" series, the All-In podcast delves deep into the crucial components of the artificial intelligence (AI) ecosystem. Featuring industry leaders Jim Latinsky, Steve from TSMC, Chase Lockmiller, Jensen Huang of NVIDIA, and Lisa Su of AMD, the episode explores the intertwined relationships between rare earth materials, semiconductor manufacturing, AI infrastructure, and public-private partnerships shaping the future of technology and national security.

MP Materials and the Rare Earth Supply Chain

Jim Latinsky, CEO of MP Materials, kicks off the discussion by sharing his journey from managing a successful hedge fund to leading MP Materials, which is now the sole supplier and refiner of rare earth materials in the United States. Chamath Palihapitiya commends Latinsky's transformation, highlighting MP Materials' pivotal role in the AI supply chain.

Jim Latinsky [00:24]: "We're 100% of the American industry."

Latinsky emphasizes the significance of rare earth magnets as the "feedstock to physical AI," underscoring their necessity in robotics, drones, and other AI-driven technologies. He details the complexities of mining and refining rare earths, a process dominated by China, and explains MP Materials' strategic initiatives to secure and expand the supply chain domestically.

In recent developments, Latinsky announces two major partnerships:

1. Public-Private Partnership with the Department of Defense (DoD):

   • Investment: $400 million
   • Purpose: Provide a price floor for rare earth commodities to counteract Chinese mercantilism, ensuring sustainability and growth of the U.S. rare earth industry.
   • Impact: Accelerate the magnetic supply chain and expand production capacity with the DoD as a strategic partner.

   Jim Latinsky [04:20]: "This is a true win-win... great for MP shareholders and national security."

2. Partnership with Apple:

   • Objective: Expand magnetic manufacturing capabilities to meet Apple's stringent specifications, securing foundational partnerships in the tech industry.

Latinsky also addresses workforce challenges in the rare earth sector, highlighting the need for skilled labor despite low graduation rates in mining-related fields.

Jim Latinsky [09:38]: "We're training people and there's absolutely talent out there. People are hungry to do it."

Onshoring Semiconductor Manufacturing: TSMC's Arizona Facility

Steve from Taiwan Semiconductor Manufacturing Company (TSMC) discusses the company's first silicon output at their Arizona facility's 2-nanometer line. He outlines the challenges and successes of establishing advanced semiconductor manufacturing on U.S. soil.

Steve [14:04]: "We're super excited about the progress in U.S. manufacturing. Where there's a will, there's a way."

Key Points:

• Workforce Development:
  TSMC faced initial hurdles in hiring and training a qualified workforce but achieved parity in chip yields between their Arizona and Taiwan facilities within a year.

• Cost Competitiveness:
  Manufacturing in the U.S. incurs higher costs, projected to be "low double digits" more expensive than in Taiwan. However, Steve argues that reliability and supply assurance compensate for the increased expenses.

  Jason Calacanis [16:57]: "Because it's unrealistic to think the United States could compete on cost. Am I correct?"
  Steve [16:37]: "We're going to pay a little bit more... less than 20%."

• Scaling Challenges:
  Addressing massive chip demands requires scaling the manufacturing ecosystem, with an emphasis on geographic diversity to mitigate supply chain risks.

AI Infrastructure and Data Centers: Crusoe's Ambitious Build-Out

Chase Lockmiller, CEO of Crusoe, presents a vision for the AI Industrial Revolution, focusing on the exponential growth of AI infrastructure and the accompanying energy demands.

Key Highlights:

• AI Factories:
  Crusoe builds "AI factories"—data centers designed to produce intelligence by processing vast amounts of data and algorithms.

  Chase Lockmiller [31:30]: "We're building AI factories at scale and speed... a gigawatt scale computer to drive human progress forward."

• Energy Consumption:
  AI-driven data centers are projected to consume 10% of U.S. power by 2030, necessitating massive investments in energy infrastructure.

• Modular Construction:
  Utilizing modular components, Crusoe rapidly scales its facilities, exemplified by their gigawatt-scale cluster in West Texas.

• Job Creation and Workforce:
  The AI infrastructure boom will be the largest job creation catalyst in history, requiring millions of skilled workers to build, operate, and maintain these facilities.

• Public-Private Collaboration:
  Partnerships with energy companies like Redwood Materials and GE Vernova underline the necessity of integrated efforts to support AI growth.

Chase Lockmiller [39:28]: "AI infrastructure will be the largest job creation catalyst that we've ever seen."

NVIDIA's Perspective on AI and Semiconductor Leadership

Jensen Huang, CEO of NVIDIA, provides insights into the company's role in advancing AI technologies and the broader semiconductor industry.

Key Insights:

• AI as a Productivity Booster:
  AI enhances productivity across all sectors, enabling companies to execute more ideas efficiently.

  Jensen Huang [47:00]: "AI is creating jobs. It causes us to be able to create things that other people would like to buy."

• AI as an Equalizer:
  AI democratizes capabilities, allowing even non-experts to harness powerful tools.

  Jensen Huang [47:00]: "AI is the greatest technology equalizer of all time."

• Competitive Landscape:
  While China is heavily investing in AI and semiconductor technology, Huang remains optimistic about U.S. leadership due to robust innovation pipelines and strategic collaborations.

  Jensen Huang [53:42]: "Onshoring next generation manufacturing is going to be insanely technology driven..."

• Supply Chain Resilience:
  Emphasizes the importance of an open ecosystem and collaboration across public and private sectors to maintain competitiveness against global rivals.

• Future of AI Infrastructure:
  Predicts a multi-trillion-dollar investment in AI infrastructure annually, essential for sustaining AI-driven industries akin to energy and the internet.

Jensen Huang [51:46]: "We're probably a couple of hundred billion dollars, maybe a few hundred billion dollars into a multi-trillion dollar infrastructure build out per year."

AMD's Role in the AI Ecosystem

Lisa Su, CEO of AMD, engages in discussions about the semiconductor industry's capacity to meet growing AI demands and the strategic importance of onshoring.

Key Points:

• First Silicon Output at TSMC Arizona:
  Lisa Su inquires about the progress and challenges faced by TSMC's Arizona facility, touching on workforce qualifications and scalability.

  Lisa Su [14:36]: "We have to make sure that there's a lot of geographic diversity and capability."

• Competitive Edge:
  Emphasizes AMD's commitment to advancing semiconductor manufacturing to support AI advancements, ensuring that the U.S. remains a pivotal player in the global tech landscape.

• Public-Private Partnerships:
  Advocates for collaborative efforts between the government and private sector to maintain and enhance technological supremacy.

Public-Private Partnerships and National Security

Throughout the episode, the critical role of public-private partnerships emerges as a recurring theme, particularly in securing supply chains and fostering innovation essential for national security and economic growth.

Jim Latinsky [11:07]: "This administration did something totally unique... this deal was led by DoD."

These collaborations are depicted as essential in countering global competitors, particularly China, by ensuring that strategic industries like rare earths and semiconductors remain robust and independent.

Future of AI and Manufacturing

The conversation moves towards the future trajectory of AI and manufacturing, with experts predicting transformative changes across various industries:

• Physical AI vs. Data Center AI:
  Predictions indicate that physical AI, embodied in robotics and autonomous systems, will become a significant market within five years, complementing the existing data center-centric AI deployments.

  Lisa Su [21:47]: "I think it becomes a significant end market."

• AI-Driven Manufacturing:
  Envisions factories orchestrated by AI, seamlessly integrating robotics and intelligent systems to enhance productivity and innovation.

  Jensen Huang [57:38]: "Every industrial company will be an AI company or you're not going to be an industrial company."

• Energy Infrastructure:
  Emphasizes the necessity of scaling energy production to support the burgeoning AI infrastructure, highlighting the parallels with historical energy investments that fueled industrial growth.

Conclusion

The episode culminates in a robust discussion about maintaining and advancing American competitiveness in the global AI race. With insights from industry leaders like Jensen Huang and Lisa Su, the podcast underscores the importance of strategic partnerships, investment in workforce development, and sustained innovation to ensure the U.S. not only keeps pace but leads in the AI-driven future.

Chamet Palihapitiya [56:06]: "We have to continue to advance it. Onshoring next generation manufacturing is going to be insanely technology driven."

The collaborative efforts between government entities like the Department of Defense and visionary companies ensure that the United States remains at the forefront of the AI revolution, leveraging its unique advantages in innovation, resources, and strategic leadership.

Notable Quotes:

• Jim Latinsky [00:24]: "We're 100% of the American industry."
• Jim Latinsky [04:20]: "This is a true win-win... great for MP shareholders and national security."
• Chase Lockmiller [31:30]: "We're building AI factories at scale and speed... a gigawatt scale computer to drive human progress forward."
• Steve [14:04]: "We're super excited about the progress in U.S. manufacturing. Where there's a will, there's a way."
• Jensen Huang [47:00]: "AI is the greatest technology equalizer of all time."
• Jensen Huang [54:48]: "Onshoring next generation manufacturing is going to be insanely technology driven..."
• Chamet Palihapitiya [56:06]: "We have to continue to advance it. Onshoring next generation manufacturing is going to be insanely technology driven."

This comprehensive discussion highlights the intricate nexus between material supply chains, semiconductor manufacturing, AI infrastructure, and strategic collaborations essential for securing leadership in the global AI arena.