Jensen Huang: Extreme co-design is essential for modern computing, the slowing of Moore’s Law demands innovation, and Nvidia’s strategic shift to AI reflects market adaptation | Lex Fridman Podcast

NVIDIA's strategic pivot from gaming to AI has revolutionized the tech industry and computing landscape.

Listen on Lex Fridman Podcast

Share

Add us on Google by Editorial Team Mar. 23, 2026

Key takeaways

• Extreme co-design is necessary for solving modern computational problems that exceed the capabilities of a single computer.
• Distributed computing involves breaking down algorithms to tackle complex problems and achieve speedup.
• Moore’s Law has slowed down due to the deceleration of Dennard scaling, affecting future computational advancements.
• Designing a computer requires an understanding of the entire stack, from hardware to software.
• Nvidia’s transition from gaming GPUs to AI reflects strategic adaptation to market demands.
• The market size influences R&D capacity, which affects a company’s impact in computing.
• The introduction of fp32 into shaders was pivotal for GPU programmability.
• Implementing CUDA on GeForce was a strategic risk that proved to be a brilliant move.
• A large installed base is crucial for attracting developers to a computing platform.
• Despite criticisms, the x86 architecture remains the defining architecture of today.
• Nvidia’s strategic decisions have been key to its evolution and success in the tech industry.
• Understanding computational needs and market dynamics is essential for innovation in technology.
• The balance between specialization and generalization is crucial in technology development.

Guest intro

Jensen Huang is the co-founder, president, CEO, and board member of NVIDIA. He invented the GPU in 1999, sparking the PC gaming market and igniting the era of modern AI. Under his leadership, NVIDIA has become the world’s most valuable company and the engine powering the AI computing revolution.

Why extreme co-design is necessary

• Extreme co-design addresses the limitations of single computer solutions for modern problems.

• The reason why extreme co design is necessary is because the problem no longer fits inside one computer to be accelerated by one gpu.

  — Jensen Huang

• This approach involves integrating multiple components to solve complex computational challenges.
• Understanding the shift from single GPU solutions to multi-component systems is crucial.
• The need for extreme co-design highlights a fundamental change in computational problem-solving.
• Integrated system design is essential for addressing modern computational needs.
• Extreme co-design emphasizes collaboration across different expertise areas.
• This approach is critical for innovation in technology and computing.

Challenges of distributed computing

• Distributed computing requires breaking down algorithms to tackle complex problems.

• Now all of a sudden when you distribute the problem this way not just scaling up the problem but you’re distributing the problem then everything gets in the way… it’s just a massively complex computer science problem.

  — Jensen Huang

• A multifaceted approach to optimization is needed to overcome these challenges.
• Understanding Amdahl’s Law is important for scaling distributed computing systems.
• The complexities of distributed computing highlight the need for innovative solutions.
• Addressing multiple complex problems is essential for achieving significant speedup.
• Distributed computing involves significant computational challenges.
• This approach requires a deep understanding of computer science principles.

The slowing of Moore’s Law

• Moore’s Law has largely slowed due to the deceleration of Dennard scaling.

• We scale up linearly or we scale up based on the capabilities of Moore’s Law which has largely slowed because Dennard scaling has slowed.

  — Jensen Huang

• This trend impacts expectations for future computational advancements.
• Understanding Moore’s Law and Dennard scaling is crucial for technology development.
• The slowing of Moore’s Law affects semiconductor technology and performance improvements.
• This trend highlights the need for new approaches to computational advancements.
• The impact of Moore’s Law’s slowing is significant for the tech industry.
• Future computational advancements require innovative solutions beyond Moore’s Law.

Designing a comprehensive computer stack

• Designing a computer requires an understanding of the entire stack from hardware to software.

• when you’re designing a computer you have to have operating system of computers… you constantly have an eye on the entire stack and you’re having to do like intense discussions about the design of the entire stack

  — Jensen Huang

• Collaboration across different expertise areas is essential for computer design.
• A holistic approach is necessary for successful computer design.
• Understanding the complexity of computer design is crucial for innovation.
• Expertise in various domains is important for designing a comprehensive computer stack.
• Computer design involves intense discussions and collaboration.
• This approach highlights the importance of a comprehensive understanding of technology.

Nvidia’s strategic evolution

• Nvidia’s evolution from a gaming GPU company to an AI factory reflects strategic adaptation.

• We started out as an accelerator company… the problem is when you become a computing company it’s too general purpose and it takes away from your specialization.

  — Jensen Huang

• This shift highlights the trade-offs between specialization and generalization in technology.
• Understanding Nvidia’s transition is important for grasping its strategic decisions.
• The company’s strategic shift has been key to its success in the tech industry.
• Nvidia’s evolution reflects a response to market demands and opportunities.
• Specialization versus broader computing capabilities is a critical consideration.
• This evolution underscores the importance of strategic adaptation in technology.

Market dynamics and R&D capacity

• The market size dictates R&D capacity, influencing a company’s impact in computing.

• The problem is the market size also dictates your R and D capacity and your R and D capacity ultimately dictates the influence and impact that you can possibly have in computing.

  — Jensen Huang

• Understanding market dynamics is crucial for innovation in technology.
• R&D capacity is a critical factor in a company’s ability to innovate.
• The relationship between market conditions and R&D capacity affects tech development.
• This insight highlights the importance of strategic planning in technology companies.
• Market dynamics play a significant role in shaping a company’s influence.
• R&D capacity is essential for driving technological advancements.

Pivotal advancements in GPU technology

• The introduction of fp32 into shaders was a pivotal moment for GPU programmability.

• the second thing that we did was we created we put fp32 into our shaders that fp32 step ieee compatible fp32 was a huge step in the direction of computing

  — Jensen Huang

• This advancement enabled broader software compatibility with GPUs.
• Understanding the significance of fp32 is important for grasping GPU evolution.
• The impact of fp32 on software development was significant.
• This technical advancement was crucial for the growth of GPU technology.
• The introduction of fp32 reflects a strategic decision in GPU development.
• This advancement highlights the importance of innovation in technology.

Strategic risks and rewards in tech

• Implementing CUDA on GeForce was a strategic risk that proved to be a brilliant move.

• that was the first strategic decision that that is as close to an existential threat… it turned out to be one of the most incredibly brilliant decisions ever made by a company

  — Jensen Huang

• This decision highlights the long-term impact of strategic shifts in tech.
• Understanding the risks and rewards of strategic decisions is crucial for tech companies.
• The competitive landscape in computing involves significant strategic risks.
• This move underscores the importance of bold decision-making in technology.
• The success of CUDA on GeForce reflects a calculated risk in tech development.
• Strategic risks can lead to significant rewards in the tech industry.

Importance of a large installed base

• The installed base of a computing platform is crucial for attracting developers.

• The installed base is in fact the single most important part of an architecture… installed base defines an architecture not everything else is secondary.

  — Jensen Huang

• A large installed base is essential for the adoption of new computing architectures.
• Developer engagement is a key factor in the success of computing platforms.
• Understanding the significance of an installed base is important for tech companies.
• This insight highlights the fundamental principle of platform success.
• A large installed base drives innovation and adoption in technology.
• The success of computing architectures depends on developer engagement.

Resilience of the x86 architecture

• Despite criticisms, the x86 architecture remains the defining architecture of today.

• No architecture has ever attracted more criticism than the x86… yet it is the defining architecture of today.

  — Jensen Huang

• Understanding the resilience of x86 is important for grasping its significance.
• The x86 architecture has proven to be a critical component of modern computing.
• This statement reflects a strong viewpoint on the importance of x86.
• The resilience of x86 highlights its impact on the computing landscape.
• Despite challenges, x86 continues to be a dominant force in technology.
• The defining nature of x86 underscores its historical and current significance.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.