Bloomberg Intelligence Podcast

Episode: Bloomberg Tech Special: Nvidia CEO Jensen Huang on Upbeat Forecast
Date: November 20, 2025
Host(s): Scarlet Fu, Paul Sweeney
Guest: Jensen Huang, CEO of Nvidia

Overview

This episode features an in-depth interview with Nvidia CEO Jensen Huang, focusing on Nvidia's record-breaking forecasts, product roadmap (including Blackwell and Vera Rubin chips), global supply chain and energy constraints, the complexities of selling to China amid regulatory restrictions, relationships with major cloud and AI players (including OpenAI and Anthropic), and industry concerns around product longevity and depreciation. Insights on Nvidia’s unique approach to architecture and its strategy for global market leadership are prominent throughout.

Key Discussion Points & Insights

A. Phenomenal Demand for Blackwell; Supply Chain Strength

• [01:33–03:01]
  • Nvidia’s Blackwell GPUs are sold out in the cloud; demand is described as “off the charts.”
  • Jensen emphasizes the strength and preparedness of Nvidia’s extensive supply chain, citing key partners like TSMC, SK Hynix, Micron, Samsung, Foxconn, Quanta, Wiztron.
  • Nvidia has a robust pipeline both for Blackwell and the upcoming Vera Rubin chip.
  • Quote [01:58]:
    “We have the largest supply chain in the world. Our partners... are doing a fantastic job supporting us and all of our systems partners... business is very, very strong.”

B. The Road Ahead for Vera Rubin

• [03:01–04:47]

  • Vera Rubin is progressing through engineering stages; launch is targeted for Q3 of the following year.

  • Seven variants of the chip are in the labs, with a large engineering task force.

  • The new “rack scale architecture” (starting with Grace Blackwell) brings seamless, revolutionary data center integration.

  • Quote [03:14]: “Probably a couple of 20,000 people are working on bringing up Vera Rubin...this bring up is going beautifully. We’re on track to deliver Vera Rubin about Q3...”

  • Quote [03:54]: “This rack architecture... is incredibly complex... transitioned to Grace Blackwell Ultra is incredibly seamless. The same rack scale architecture is going to be used for Vera Rubin.”

C. Nvidia’s Position in China and Regulatory Landscape

• [04:47–06:51]

  • Nvidia currently forecasts zero sales to China due to regulatory restrictions (“our forecast for China is zero”).
  • Despite this, China remains a crucial market for both Nvidia and the U.S., and resumption of sales is a company aspiration.
  • Ongoing dialogue with U.S. and Chinese authorities continues, but for now, business expectations in China are set at zero.
  • Quote [05:24]: “Our forecast for China is zero. All of our forecast guidance that we showed, zero... China is a very important market to us... it's great for the American people that we’re able to compete in the Chinese market.”

• [06:51–08:15]

  • U.S. Commerce Department is now allowing Nvidia to export up to 35,000 Blackwell chips each to Saudi and UAE partners, with strict controls against “diversion” to China.
  • Nvidia claims to have historically managed such controls rigorously, using technology and compliance processes.
  • Quote [07:28]:
    “We’ve repeatedly tested and sampled data centers around the world and found no diversion... This is an area that we’ll continue to be rigorous on.”

D. Industry-Wide Constraints: Energy And Expansion Challenges

• [08:15–10:45]
  • Nvidia’s rapid growth (60% year-over-year, $10 billion quarter-to-quarter) creates immense sourcing and operational challenges — not just in chips but energy, land, and power.
  • Nvidia’s global network of partners and customers provides flexibility in locating resources and mitigating constraints.
  • A significant market advantage is Nvidia’s universality: “Our architecture literally runs every model.”
  • Simultaneous support for top AI model providers (OpenAI, Anthropic, XAI, Gemini, etc.) is a unique capability.
  • Quote [08:36]: “When you're growing at the rate and scale of Nvidia... everything’s a challenge. Which is the reason why Nvidia has to be world class at our supply chain... But also working downstream to work with energy providers... so that we could make sure... land, Power and shell will be ready for us.”

E. Ensuring Partner Commitments and Credibility

• [10:45–13:02]
  • Visibility into demand and alignment with the financing capabilities of strategic partners (e.g., OpenAI, Anthropic) is key before scaling up supply.
  • Nvidia maintains discipline in investments and deployments given the scale and speed of its partners’ growth.
  • AI demand is described as moving “exponentially,” with Nvidia working hard to optimize for current and future uses.
  • Quote [11:08]: “The ambition is large, but the execution is disciplined. And that's really, really important to recognize. ... OpenAI, Anthropic, these are the fastest growing company in the history of humanity. Their offtake, their end market demand is absolutely real...”

F. Depreciation, Chip Longevity, and Versatility

• [13:02–15:29]

  • Jensen argues Nvidia's GPUs (like the Ampere A100, now six years old) remain highly valuable and widely deployed, due to continual software and architectural updates.

  • Nvidia’s diversity and versatility (via CUDA) mean chips don’t suffer the same rapid depreciation as competitors’ more specialized accelerators.

  • GPUs retain relevance far beyond their official release cycle, sustaining their value and utility.

  • Quote [13:31]: “Nvidia's architect... unlike any other accelerator. And the reason for that is because of CUDA's diversity of capability and versatility... we run every single model.”

  • Quote [14:49]: “A100 is six years old. However, it is still an order of a magnitude faster than any CPU could put to bear. So it is still the best processor for much of the workload in the cloud.”

G. Nvidia’s Contribution to AI Infrastructure: Cost Clarification

• [15:29–16:35]

  • On the cost composition of a 1 GW data center: Nvidia hardware can account for “billions of dollars” (not percentages).

  • For the Vera Rubin system, of the $50–55 billion total, Nvidia’s share is about $35 billion.

  • Quote [15:57]: “Billions of dollars. ... for our Vera Rubin system, a 1 gigawatt data center is probably something along the lines of 50 to 55 and various contribution is probably about 35 of that.”

    • Jensen further underscores the distinction between percentage and absolute dollar impact.

Notable Quotes & Memorable Moments

• On Nvidia’s Architecture and Value Proposition:
  “We're the only architecture in the world that does [pre-training, post-training, and inference] fantastically. ... Most agentic systems, most clouds are running so many different diverse type of models... For all the different fields of science, Nvidia could be used across the entire lifespan of the technology.” (Jensen Huang, 13:31–14:17)

• On the Growth of AI Compute Demand:
  “You're seeing an exponential growth in the amount of compute demand necessary for AI. ... The number of applications that are going to be using these AIs is also growing.” (Jensen Huang, 12:43–13:02)

• On China Sales Outlook:
  “Our forecast for China is zero... we’re going to continue to engage the US government, continue to engage the China government, to advise them and to encourage them to allow us to go back and compete in the open market.” (Jensen Huang, 05:24–06:51)

Timestamps for Important Segments

• 01:33 — Introduction of Jensen Huang, discussion starts
• 01:58 — Demand and supply chain for Blackwell and future chips
• 03:14 — Vera Rubin engineering and architecture
• 05:24 — China forecast and regulations
• 07:28 — U.S. export controls and compliance
• 08:36 — Energy and infrastructure constraints
• 11:08 — Ensuring financing & demand alignment with top AI companies
• 13:31 — Depreciation and longevity of Nvidia chips
• 15:57 — Cost structure of AI datacenter builds

Conclusion

This episode delivers a rare executive-level look at Nvidia’s multipronged strategy for maintaining leadership in the AI hardware market, their risk management in a geopolitically complex environment, and how their technology underpins the current AI revolution. Huang’s emphasis on supply chain readiness, architectural flexibility, disciplined investment, and constant engagement with regulators and partners paint a picture of a company powering—and shaping—the exponential trajectory of AI’s global impact.