Podcast Summary

Episode Details

• Podcast: Azeem Azhar's Exponential View
• Episode: Inside the US–China decoupling: What’s really at stake (AI, rare earths, Taiwan, trade)
• Date: October 23, 2025
• Host: Azeem Azhar
• Guest: Jordan Schneider (ChinaTalk, Substack writer, US–China tech analyst)

Main Theme

This episode explores the ongoing and complex process of US–China "decoupling," particularly as it relates to high technology, rare earths, trade, and the broader global geopolitical landscape. Azeem and Jordan deconstruct the current state of US–China relations, the limitations of existing policy tools, the parallel development of technological ecosystems, and the deep entanglements that still bind these two superpowers. They also examine how AI sovereignty, chip manufacturing, and deployment strategies are reframing competition and resilience in both countries.

Key Discussion Points and Insights

The Nature of US–China Decoupling

• Decoupling is not mere separation but the development of "two parallel tech universes" that remain deeply entangled (01:26).
• The instruments for managing this decoupling—export controls, tariffs, sanctions—are outdated for the modern era (01:26).

Quote

"It's not so much separation, but the emergence of these two parallel tech universes. But they remain deeply entangled."
—Azeem Azhar (01:26)

Historical Phases of US–China Relations

• 1949–1973: Near-total isolation between the US and China.
• 1973–2013: Period of engagement and mutual economic and technological entanglement, arising from strategic interests during the Cold War and later economic liberalization (05:03).
• 2013–Present: Shift to a framing of China as a competitor (or enemy), marked by trade wars, COVID-19-era restrictions, tightening export controls, and escalating policy moves by both countries (05:03–07:04).

Quote

"It's a big mess because these are the two largest economies in the world and they're not separate."
—Jordan Schneider (07:04)

Unintended Consequences of "Control"

• Attempts at control (like export controls) often foster greater resilience in the opposing party. China's drive toward technological self-reliance is reinforced, not deterred, by US actions (07:04–08:50).
• Historical analogies abound: 1930s autarky, OPEC oil weaponization, and modern "weaponization of dependency" (13:45).

Quote

"One often meets one's fate on the path one takes to avoid it... what you may end up doing is making the other side much, much more resilient."
—Azeem Azhar (08:12)

Self-Reliance, Sovereignty, and Global Competition

• China: Historical trauma and fear of dependency drive longstanding self-reliance efforts. Recent Western moves are echoing this mindset (10:00–13:45).
• West: Movement towards "strategic autonomy" in the EU, US re-industrialization, rising concern about rare earth dependencies (12:36).
• Sovereignty: The concept is increasingly popular, but only China and the US have the depth to make it meaningful. Other regions (like the EU) have ambitions but face structural limitations (21:27–22:50).

Quote

"Xi has this great line, I think he said it in a 2020 speech where he's like, 'we want to not be dependent on other countries and we want other countries to be dependent on us.'"
—Jordan Schneider (19:47)

The AI and Chip Supply Chain

• AI Sovereignty: Only China and the US can (roughly) control the full stack from chips to clouds to models (23:41–25:26).
• Chips: Europe’s talk of semiconductor sovereignty is unrealistic given the scale and tech lead of US/Asian firms (22:09–22:50).
• Open Source and Deployment: China excels in open source models and rapid deployment, but lacks cutting-edge chips. The US leads in frontier AI models (22:50–29:21).

Notable Moment

"Are you really going to try to spend $50 billion, $75 billion to build a competitor to Intel or TSMC or Samsung? ...the reality is like no, you're still going to get all of your advanced chips for decades to come from Asia or from the U.S."
—Jordan Schneider (22:09)

• There's more fluidity in the AI arena above the chip layer, leaving space for new entrants (Abu Dhabi, open-source providers), but entrenched US and Chinese players dominate the critical hardware and model infrastructure (27:13–28:29).

Deployment versus Frontier in AI Strategy

• China: State-backed emphasis on widespread deployment across industries, aiming for transformative impact via adoption rates (31:12).
• US: Continues to push the frontier of AI research and development.
• Broad deployment might, in fact, be a greater determinant of economic impact than first-mover breakthroughs (32:09).

Quote

"Jeff [Ding] had his book recently and he made the case that it's less about breakthrough innovation and more about broad diffusion and deployment."
—Azeem Azhar (31:12)

Societal Adaptation and Tech Acceleration

• Both societies face challenges in adapting to rapid technological change. A key variable will be how each system manages disruptive impacts—especially in employment and social stability (33:36–35:32).
• Historical episodes like the 1980s protests in China illustrate the risks of change outpacing social management capacities.

Quote

"Which societies will hold up better in a time of sort of rapid employment turnover is one that I find fascinating and one that I think is a very open question."
—Jordan Schneider (33:36)

Lessons from the Chinese Development Model

• China’s "performance legitimacy" (success through results, not just procedures) is setting a new standard globally for what state-led modernization can deliver (35:32–36:54).
• Still, Jordan cautions: modernity's "dark sides"—e.g., repression in Xinjiang—shouldn't be ignored (36:54).

Quote

"If we're calling China the archangel of modernity, we also have to understand that there are parts of that modernity which I hope listeners of this are very uncomfortable with..."
—Jordan Schneider (36:54)

Understanding China from the Outside

• While Chinese elite politics remain opaque, industrial strategies (e.g., in EVs) are surprisingly legible to those who research carefully (39:48–41:33).
• Humility and careful observation, rather than defeatism, are the right approach.

Memorable Final Segment: Lessons from the Manhattan Project

• Jordan keeps a Manhattan Project poster as a reminder of the gravity of technological change and the risks in US–China dynamics (42:06).
• Ultimately, he sees his work as contributing to reducing the risk of great power war (42:53).

Quote

"The point of ChinaTalk and the work I've been doing for the past eight years is actually aimed at avoiding great power war."
—Jordan Schneider (42:53)

Looking to 2030: What Would Progress Look Like?

• Preventing War Over Taiwan: The US must clearly signal that conflict would not benefit China (43:13).
• Mutual Rethinking: China needs to recognize that engagement with the world need not be confrontational, but deep-seated historical anxieties make this difficult (43:13–44:50).

Quote

"America needs to make it really clear...that a war in Taiwan is a war that China would lose."
—Jordan Schneider (43:13)

Notable Quotes & Timestamps

• "It's not so much separation, but the emergence of these two parallel tech universes. But they remain deeply entangled."
  —Azeem Azhar (01:26)

• "It's a big mess because these are the two largest economies in the world and they're not separate."
  —Jordan Schneider (07:04)

• "One often meets one's fate on the path one takes to avoid it... what you may end up doing is making the other side much, much more resilient."
  —Azeem Azhar (08:12)

• "Xi has this great line... 'we want to not be dependent on other countries and we want other countries to be dependent on us.'"
  —Jordan Schneider (19:47)

• "Are you really going to try to spend $50 billion, $75 billion to build a competitor to Intel or TSMC or Samsung? ...the reality is...no, you're still going to get all of your advanced chips for decades to come from Asia or from the U.S."
  —Jordan Schneider (22:09)

• "Jeff [Ding] had his book recently and he made the case that it's less about breakthrough innovation and more about broad diffusion and deployment."
  —Azeem Azhar (31:12)

• "If we're calling China the archangel of modernity, we also have to understand that there are parts of that modernity which I hope listeners of this are very uncomfortable with..."
  —Jordan Schneider (36:54)

• "The point of ChinaTalk and the work I've been doing for the past eight years is actually aimed at avoiding great power war."
  —Jordan Schneider (42:53)

• "America needs to make it really clear...that a war in Taiwan is a war that China would lose."
  —Jordan Schneider (43:13)

Timestamps for Key Segments

• Introduction & Podcast Partnership: 00:00–01:26
• US–China Tech Decoupling Framing: 01:26–05:03
• History & Periodization of Relations: 05:03–07:04
• The Resilience Paradox, Rare Earths: 07:04–08:50
• Self-Reliance, Autarky Analogies: 08:50–13:45
• Weaponization of Dependency: 13:45–15:42
• Policy Coherence: 15:42–18:13
• System Building, Innovation Systems: 18:13–19:47
• Sovereignty & AI Stack Reality: 21:27–23:41
• Cloud, Chips, Open AI Models: 23:41–28:29
• Deployment vs. Frontier AI: 28:29–32:09
• Techno Accelerationism & Societal Adaptation: 32:09–35:32
• Performance Legitimacy, Modernity: 35:32–38:53
• Legibility of Modern China: 38:53–41:33
• Manhattan Project Lessons: 42:06–42:53
• Averting War, Future Scenarios: 43:13–44:50

Conclusion

This episode offers a nuanced, historical, and pragmatic analysis of the US–China decoupling, emphasizing the persistence of interdependence, the limits of state power, and the messy, unpredictable path of exponential technological change. Both leading powers are improvising, their policy tools often mismatched for the scale and complexity of 21st-century rivalry. Yet, the conversation ends on a hopeful note: recognizing the stakes, learning from each other, and working—however imperfectly—toward a more stable and peaceful future.