ChinAI #308: Runaway Tech Capital AI vs. Socialist Open-Source AI?
The Beijing Cultural Review's view on U.S.-China Competition in AI
Greetings from a world where…
laughter is carbonated holiness — Anne Lamott
…As always, the searchable archive of all past issues is here. Please please subscribe here to support ChinAI under a Guardian/Wikipedia-style tipping model (everyone gets the same content but those who can pay support access for all AND compensation for awesome ChinAI contributors).
Feature Translation: How has the West’s misjudgment of China’s AI ecosystem distorted the global technology competition landscape
Context: Across almost 20 rounds of voting for Around the Horn issues, I’ve never seen a blowout victory like the one you all delivered for this week’s feature translation: the Beijing Cultural Review [文化纵横] editorial board’s commentary on the two paths forward for AI development: 1. One pushed by the U.S.’s runaway tech capital; 2. DeepSeek’s “human-centered” approach grounded in open-source principles. The piece does not hold back. Here’s one particularly spicy quote:
Right-wing capital in Silicon Valley, represented by figures like Peter Thiel, play up the narrative of U.S.-China competition to elevate its own importance, thereby securing its own interests and technological edge. In doing so, it has pushed for tech export controls on China and deregulation at home, aggressively advancing the deployment of new technologies such as AI and cryptocurrency, and embracing an “accelerationist” path with little regard for the broader welfare of humanity.
In my opinion, this piece makes some overly crude comparisons. And I’ve written in the past about how Chinese thinkers misperceive U.S. technology strategy (see ChinAI #292 on the misperception spiral in U.S.-China tech policy). Still, Beijing Cultural Review represents an influential platform for Chinese liberal intellectuals, so it’s useful to understand how they see things.
***Thanks to Marianne Lu for contributing this week’s feature translation and analysis. Marianne is a technology and security policy fellow at RAND. She recently graduated from Stanford University, where her thesis focused on U.S.-China AI competition. What follows is her analysis (lightly edited by me).
Key Takeaways: The authors situate DeepSeek’s rise within a global AI ecosystem long dominated by the U.S.’ “monopolistic tech capital.” In pursuit of profit and technological advantage, they argue, American AI developers and their backers have played up the narrative of U.S.-China rivalry, pushed for export controls, and accelerated domestic AI deployment with little concern for broader societal welfare. DeepSeek’s emergence thus “marks a historic event” in three important ways:
It disrupts the narrative that technological progress depends on huge amounts of capital: “With relatively modest resources, it achieved impressive technological breakthroughs—strong evidence that innovation is not determined by volume of capital, and that technological progress is not the exclusive domain of monopolies.”
Its open source models have paved the way for a more inclusive model of technological development: “By enabling affordable, localized deployment, it has rapidly entered sectors like education, healthcare, and manufacturing––while alleviating widespread concerns around data privacy and security.”
It has changed the rules of the game in the AI industry: “Major tech companies have been forced to reconsider their traditional strategy of maintaining dominance through intellectual property and capital advantages.”
Beyond its immediate technological contributions, DeepSeek also prompts a broader reassessment of the relationship between technology, society, and power.
The piece raises concerns about the erosion of labor, ethics, and social bonds in an age of AI automation, one in which humanity risks losing its connection to the real world.
In particular, it argues that the “profit-driven impulses of capital” are steering humanity into a future marked by “the online violence and echo chambers caused by unconstrained individuals on the internet, the democratic decline and political dysfunction unfolding in the West, and the possibility of a new Luddite movement now brewing.”
The authors nevertheless urge against pessimism, arguing that Deepseek––and the socialist system behind it––can help forge new, socially beneficial paradigms for AI.
Not only does DeepSeek present a new technological path, but the authors are also optimistic that “collective action under socialism” can help safeguard social institutions and everyday life.
The final paragraph encapsulates the piece’s core thesis: “In March 2025, OpenAI lobbied the U.S. government to promote its concept of ‘democratic AI’ and to ban what it labeled as ‘authoritarian AI’ from China. Yet it is already clear which path has been captured by runaway tech capital, and which is seeking a model of AI that is compatible with society.”
FULL TRANSLATION: How has the West’s misjudgment of China’s AI ecosystem distorted the global technology competition landscape
ChinAI Links (Four to Forward)
Should-read: China is trying to create a national network of cloud computing centers
An in-depth analysis of about the role of the Eastern Data Western Compute 东数西算 project in China’s vision for a “national network of computing.” This is from the Sinocities substack on China’s urbanization and infrastructure development, written by Andrew Stokols, postdoctoral research fellow at MIT’s Department of Urban Studies and Planning. H/t to Mary Clare McMahon for sharing.
Should-read: Translated Notes on Tencent’s NDR
Rui Ma, editor at Tech Buzz China, translated and summarized the notes from one of the articles from last week’s Around the Horn Issue. One of the points on Tencent’s cloud and GPU strategy from her Twitter thread: “Tencent Cloud’s priority is internal AI use, not external GPU leasing (unlike Alibaba Cloud).”
Should-read: The free‑living bureaucrat
Michael Lewis profiles Heather Stone of the Food and Drug Administration, as part of The Washington Post’s essential series on the people at the heart of the U.S. federal government. I took detailed notes on Lewis’s other piece in this series on Chris Mark of the Department of Labor in this ChinAI issue from last October.
Should-read: A Defense of U.S. Education as an Export
Chris Blattman, a UChicago political scientist, posted a thought-provoking thread that started with this: “Try to imagine a US President telling Ford that they shouldn't sell cars to foreigners, and that they should cut their R&D. It wouldn't happen. But that's what we're doing to one of America's most important exports: Education.”
Thank you for reading and engaging.
These are Jeff Ding's (sometimes) weekly translations of Chinese-language musings on AI and related topics. Jeff is an Assistant Professor of Political Science at George Washington University.
Check out the archive of all past issues here & please subscribe here to support ChinAI under a Guardian/Wikipedia-style tipping model (everyone gets the same content but those who can pay for a subscription will support access for all).
Also! Listen to narrations of the ChinAI Newsletter in podcast format here.