America and China have different attitudes about and approaches to AI and work

Ross Douthat, Why China Isn’t Worried A.I. Will Replace Its Workers, NYTimes, May 14, 2026.

The United States and China are really the only two countries that matter right now in shaping the A.I. future. As President Trump and President Xi Jinping meet in Beijing, there’s a kind of Cold War atmosphere, with people talking about an A.I. arms race. But who is winning? Are we even in a race at all? Kyle Chan, a foreign policy fellow at the Brookings Institution, says it’s hard to call it a race because the U.S. and China have very different A.I. goals.

I presented the conversation to Claude (Sonnet 4.6 Adaptive) and asked it for a summary. 

 * * * * *

I have uploaded conversation to the project area, filename “China AI workers.”

It is between Ross Douthat, who writes for The New York Times, and Kyle Chan, who is a foreign policy fellow at the Brookings Institution. I would like you to summarize the conversation for me. I’m particularly interested in the Chinese attitude toward AI in general and the (proper) relationship between AI and humans.

This is a rich and wide-ranging conversation. Here's a summary, with particular attention to the themes you flagged.

Overall frame

Douthat's central premise is that the U.S. and China are in an AI "arms race," but Chan's main contribution is to complicate and largely deflate that framing. China, he argues, is running a different kind of race, not the same race at a slower pace.

The Chinese approach to AI: practical, not transcendent

The sharpest contrast Chan draws is between the American fixation on AGI — the idea of a "machine god" that surpasses all human capability — and the Chinese orientation toward diffusion and application. China wants AI in food delivery, ride-hailing, health care, government services, and factory floors. They want smaller, cheaper, more deployable models, not necessarily the most powerful frontier model. Chan puts it directly: Beijing is not "AGI-pilled."

He sees this as continuous with how China approached the internet era: not as a transformative metaphysical event, but as infrastructure to be integrated into daily life and governance. The Chinese state is also notably more willing to use open-source models, giving away AI capabilities to gain adoption at scale — both domestically and in the Global South — which is a strategy the commercially driven American labs largely ignore.

The proper relationship between AI and humans: tool, not god

This is where the conversation is most relevant to your homo ludens framework. Chan describes Beijing's view of AI as essentially instrumental — AI serves human economic and social purposes. The Chinese government is actively hostile to AI use that it considers "nonproductive": AI companions, AI boyfriends/girlfriends, even excessive gaming. The fear is that young Chinese will waste time on AI-mediated entertainment and social substitution when they should be building the economy.

There's a revealing parallel Chan draws: the same regulatory instinct that cracked down on video games and private tutoring platforms is now being pointed at AI companionship apps. The state's implicit position is that AI must serve the collective project — not become a substitute for human connection, productivity, or reproduction. This is a deeply anti-mystical, anti-transcendent view of the technology.

So the Chinese "proper relationship" between AI and humans is essentially: AI augments human labor, fills demographic gaps (falling birthrate, shrinking workforce), and extends state capacity. It does not replace human flourishing or social life; it should not become an end in itself.

The demographic driver

Chan adds an important materialist dimension: China's embrace of robotics and AI is partly driven by sheer demographic necessity. Their labor force peaked over a decade ago, youth unemployment is near 17%, and the birthrate is roughly two-thirds the U.S. rate. This produces a different calculus than in America — the question isn't whether a robot waiter displaces a human worker, but whether you have a waiter at all. AI and robotics fill a structural labor hole, which makes China more tolerant of automation than Western economies that still have more human workers to protect.

The public mood: anxiety about falling behind, not about being displaced

Where American anxiety about AI tends toward fears of job loss, surveillance, or existential catastrophe, Chan describes Chinese popular anxiety as nearly the inverse: fear of not using AI enough, of being left behind in a hyper-competitive labor market. Individual workers, companies, and policymakers all share this orientation. Chan calls this a "mirroring" — the individual's fear of losing competitive edge mirrors China's national fear of falling behind the U.S. technologically.

On AGI and existential risk: skeptical distance

Chan thinks it's telling that when Trump relaxed chip export controls, Chinese AI companies wanted the chips but Beijing declined — preferring to build domestic semiconductor capacity rather than become dependent on American supply chains. This, he argues, is not the behavior of a country sprinting toward superintelligence. A true AGI race would involve gobbling up every available chip. China is playing a longer, more measured game.

He also pushes back gently on the "race" metaphor itself: the intermediate risks (cyberwarfare, biosecurity) are, in his view, underrated and real; the AGI-superintelligence scenario is overblown. He ends by urging dialogue between the U.S. and China on shared risks from rogue actors and open-source proliferation, while acknowledging the deep mutual distrust that makes formal arms-control agreements essentially impossible at this stage.

A thought for your book

Chan's portrait of Beijing's AI worldview sits almost perfectly at the opposite pole from Silicon Valley's quasi-religious eschatology — no Omega Point, no machine god, no homo deus. What you might find generative: the Chinese model is also, in its own way, a version of homo economicus applied to AI — purely instrumental, collectively economistic, hostile to play and to what the government calls "wasted time." It's not an alternative to the Silicon Valley eschatology so much as its mirror image — the same foreclosure of genuine play, just in a different ideological register.