Bloomberg Talks

Episode: Chris Miller Talks AI Demand, Chips
Date: February 26, 2026
Host(s): Bloomberg (with Carol and John/Emily)
Guest: Chris Miller, Author of Chip War and Professor at the Fletcher School, Tufts University

Overview

This episode features a deep-dive interview with Chris Miller, renowned author of Chip War, exploring the accelerating demand for artificial intelligence (AI), the global semiconductor industry, and their geopolitical implications. The conversation spans the historic and modern cycles of chip supply and demand, where the sector stands amidst the AI boom, debates over the real economic value of massive capital expenditures, and the United States’ strategic efforts to reshape semiconductor production in response to growing dependence on Asian manufacturers.

Key Discussion Points & Insights

1. The Quantum Leap in Chip Demand for AI

• AI and Data Center Growth:
  • Chris Miller emphasizes how the explosion in AI applications has catapulted demand for semiconductors—particularly those powering data centers (01:27).
  • “The industry is supercharged, its size growing faster than ever, but dealing with some of the same supply chain choke points that it had a decade ago.” (Chris Miller, 01:27)

2. Cycles in the Semiconductor World: Changed by AI?

• Historic Cyclicality:
  • Traditional boom-and-bust cycles—characterized by periods of overinvestment and then gluts—still apply, but AI represents a ‘step change’ like smartphones did (02:36).
  • “What we're seeing right now is that type of step change, a huge increase in just the baseline amount of chips that we're going to need for data centers driven by AI.” (Chris Miller, 02:36)
• Lasting Demand:
  • Miller predicts ongoing spikes in compute requirements due to AI, leading to “a lot more of the AI chips that go inside of data centers.”

3. Will AI Deliver on Its Promises?

• Evaluating Return on AI Investment:
  • The hosts express skepticism on whether vast capital expenditures will pay off, especially for non-tech companies struggling to realize gains (03:18, 04:44).
  • “Do we live in a world where this is a layer, just like the Internet was a layer of technology?” (Co-host, 04:44)
• AI’s Impact Already Evident:
  • Miller argues AI’s potential has already materialized, noting rapid advancements post-ChatGPT that seemed impossible only years earlier (03:34).
  • “We've already had so much new capability generated just in the couple of years since ChatGPT that... it's an absurd question to ask, will AI deliver? It already has in a lot of ways.” (Chris Miller, 03:34)
• Future Returns Hinged on Widespread Embedding:
  • Miller foresees AI as a default layer in technology, drawing parallels to the Internet’s integration (05:49).

4. Profitability & R&D in the AI Era

• Current Business Models:
  • “Their margins on their inference business are, are not just positive but, but quite good. And so that I think is pretty strong evidence that the delivery of already existing AI services is a pretty profitable business.” (Chris Miller, 05:49)
• R&D Justification:
  • Despite the high costs, Miller contends it would be “hard to argue we should dramatically slow down R&D” given ongoing advancements (06:20).
  • “Would you really like to be the CEO of the only big tech company that's not investing in AI? That doesn't seem like a very comfortable place to be.” (Chris Miller, 06:53)

5. Where Should Industry Attention Focus?

• Holistic View Required:
  • Miller cautions against fixating on only one part of the semi supply chain (Nvidia, China, TSMC), emphasizing the interconnectedness of chip designers, manufacturers, and even materials suppliers (07:43).
  • “If you only look at one part of the supply chain, you miss the challenges that other parts are often facing.” (Chris Miller, 07:43)

6. Geopolitics: China, U.S. Policy & the Supply Chain

• Chip Export Restrictions as Leverage:
  • The discussion covers U.S. chip export bans to nations like China and the “arms race” analogy (08:31).
  • “The primary reason [Chinese tech companies] struggled to deploy AI products at scale is because they’ve struggled to get access to all of the computing capabilities. …This is still a very powerful card that the U.S. has to play.” (Chris Miller, 09:05)
• Industrial Policy, TSMC, and U.S. Manufacturing:
  • U.S. attempts to reshape the chip landscape—through support for TSMC, Intel, and policy tools—will not yield fast fixes (09:59-10:21).
  • “The chip industry has involved hundreds of billions of dollars of Capex over the last several decades. And so it's just not going to move fast. …You’ve got to plan for years of implementation...” (Chris Miller, 10:52)

Notable Quotes & Memorable Moments

• On AI’s transformation of the chip market:

  • “We’re just going to need a lot more compute for AI purposes in the future.” (Chris Miller, 02:36)

• On skepticism toward AI’s ROI:

  • “It seems to me that we should…be excited that there's actually companies willing to invest in the infrastructure that's going to deliver all of these capabilities.” (Chris Miller, 04:22)

• On investing in AI R&D:

  • “Would you really like to be the CEO of the only big tech company that's not investing in AI?” (Chris Miller, 06:53)

• On U.S.-China chip tensions:

  • “It does, I think, seem to me that this is still a very powerful card that the US has to play. And so I think we should be very careful or on any decisions to give China more access or at least make sure that we're getting something in exchange for that.” (Chris Miller, 09:35)

Key Timestamps

• 01:27: Chris Miller on the scale of AI-driven semiconductor growth
• 02:36: Step change in demand—AI vs. historical patterns
• 03:34: Will AI “deliver” on the promise? Retrospective and forward-looking thoughts
• 05:49: Economic viability of AI service infrastructure and inference margins
• 07:43: The complexity of the global semiconductor supply chain
• 09:05: Geopolitical considerations; US chip export restrictions and China’s challenges
• 10:21: US industrial policy: Effects of TSMC/Intel support; long-term restructuring

Conclusion

Chris Miller offers a balanced and insightful perspective on how AI’s insatiable appetite for compute power is redrawing the semiconductor landscape—magnifying old supply chain issues, altering investment calculus, and heightening global strategic competition. He stresses that while challenges remain, particularly concerning the over-reliance on Asian manufacturing and the logistical timeline of industrial policy, the industry’s trajectory toward more pervasive AI integration is both inevitable and largely positive.

This episode is a must-listen for professionals keen to understand the fundamental shifts ongoing in semiconductors, AI’s transformative impact, and the rising stakes in the global tech race.