After The AI Revolution

“From our ugliness will grow the soul of the world.”

—Andrei Platonov, “Soul”

As the race to build ever-better artificial intelligence continues, the relationship between its chief competitors, the U.S. and China, is increasingly being defined by this quest. As partners in this strange new dance, they are the only two civilizations with the necessary scale to engage — and yet, they could not be more different.

Both societies continue to struggle with deep-seated malaise in some sectors, like racism in the U.S. or rural-urban inequality in China, but the ruling classes in both countries have more fun focusing on things like AI and advanced computation instead of untangling old knots. The problems of the everyday, many of those in this ruling class seem to believe, are insoluble and can safely be ignored; after the AI revolution, everything will be fine.

As such, both societies are organizing culturally, economically, and politically around the pursuit of this new technology: China is adopting more capitalist “characteristics,” while the U.S. is embracing more tech-driven authoritarian state controls. But AI will not converge China and America into any one system. It will refract their civilizational logics into distinct technological futures: America’s restless frontier vs. China’s paternalist social engineering. Both futures make important things visible, though in radically different ways.

Artificial general intelligence — like “a country full of geniuses in a data center,” as Anthropic CEO Dario Amodei calls it — will confirm the civilizational differences between China and the U.S.; acting like a prism, it will refract these converging beams of light and send them in two separate directions. American genius seeks to trace the frontiers of knowledge; Chinese genius often finds itself in the service of a larger social goal.

Fears of a direct collision between the two countries, such as either country attacking Taiwan’s chip refineries, are likely far-fetched. The notion that Chinese and American AI are in direct competition, as evidenced by the stock market’s slide when DeepSeek was announced, is also not quite accurate. Rather, the technologies developed and adopted by each, driven in part by national competition, will be what transform the different peoples.

Both societies have their own archive of structures of thought, value systems and methods. Both are in motion toward the glowing light of the frontier — deep learning, foundational models. As they draw nearer to it, they draw nearer to each other in ethos and social structure.

Competition creates a certain symmetry. For now, China and the U.S. are both tracking toward a common goal: technological mastery enabled by the universally applicable scientific method, physics and mathematics. But once either country enters the prism — the moment when, perhaps, AGI emerges — their structural differences, like America’s deeper financial reserves and China’s manufacturing sector and state-subsidized electricity, will shape their future trajectories.

The U.S. will use, or be used by, AGI in the service of American goals and American problems: patchy regulation, a dominant private sector, a cacophony of voices pursuing ever more complex advances without pausing to ask why. We can imagine the future of American science and thought processes tracking north, bathed in blue light. China’s AGI, in contrast, colored red heads southward, toward the regulation of everyday life, an extended hand offering a generic standard of living for its masses.

Even if their technical bases are similar, the two emergent AI ecosystems will differ in function, culture and export influence. Each refracted path will reinforce itself through local policy, public expectations and industry structure (for example, by choosing to rely on open-source models or not). And for those of us who are simply passengers observing this journey, the years to come will feel like we are changing both individually and as a species, alone and in a swarm of people, almost as if by instinct. Is that what evolution feels like?

A sense of foreboding lurks behind the bravado and brittleness of American life today; this might be the last few years before an epochal crack in everyday existence. Coexisting with an Other that might be helpful but is indifferent and possibly predatory, accepting what is uncontrollable or incomprehensible — these are tendencies that do not come naturally to Americans, whose cultural habit is to organize events in an effort to master them. In China, society trundles on, automatic and without much passion, the secret engines of the future churning, building a new world in which human intelligence is neither alone nor superior.

Few in the Trump Administration, Silicon Valley or China’s planning apparatus stop to ask: What is our goal with technology? What constitutes a flourishing human life?

“Few in the Trump Administration, Silicon Valley or China’s planning apparatus stop to ask: What is our goal with technology? What constitutes a flourishing human life?”

Can AI Make Invisible China Visible?

It has become a cliché that Chinese science excels at applications and American science at frontier breakthroughs. Strictly speaking, science knows no nationality; it’s a tool. “Chinese science” is simply the way that the Chinese use that tool. So what uses do China and the Chinese Communist Party have in mind for AI?

In “Invisible China,” the American economist Scott Rozelle discusses at length the hundreds of millions of left-behind Chinese and the obstacles they face in moving to a higher level of economic or intellectual life. Even though China has made remarkable progress in eradicating absolute poverty, and even though the objects that accompany prosperity — physical infrastructures like roads, bridges, internet connectivity — are in place, a culture of poverty remains, Rozelle notes. It is a hard thing to go from illiteracy to university in two or three generations.

In its new five-year plan, China’s government employs a strategy called “AI+” to rapidly accelerate the sociological change from rural deprivation to a sort of median level of basic prosperity, basic education and healthcare. If the invisible people of China are never made visible, China will be stuck in the middle-income trap. No matter how many fancy airports and robots China has, if a huge proportion of the population is made up of peasants — not just by occupation but by cultural formation — China will never really feel “developed.”

There aren’t many doctors in China’s villages. But AI diagnosticians can make diagnoses, and locals can do the manual labor of nursing. There also aren’t enough teachers, but personalized AI tutors can work with students, while their elders, deprived of education themselves, keep the school clean. AI-driven early childhood assessment could identify cognitive and language delays earlier, triggering interventions. The endless struggle with nature’s unpredictability can now be augmented with AI-enabled precision agriculture (soil sensors, drone imagery, weather prediction) that boosts smallholder productivity.

This is how AI will make the invisible Chinese visible — not by eliminating poverty through a single breakthrough, but by integrating the rural underclass into the same informational infrastructure that already shapes urban life. These are the projects of the Chinese state, which doesn’t seek to profit from equity valuations or returns on investment, but rather sees the Chinese population as its most basic form of capital. By using AI to “upgrade” the rural population — to bring them into the fold — the Chinese government hopes to drive economic growth and invent new applications and uses for AI. All such applications, as with Chinese technologies like Huawei telecoms or BYD cars, are easily exported to the Global South.

A country is, in a sense, a large language model, and the country of China, much like its AI models, is trained on and dominated by the speech of its government. At a certain moment recently, that government undertook a global civilizational initiative. Instead of nation-states, civilizations would meet: China, the Islamic Ummah, the “Russian world,” the West (depending on how you define it). But then, what civilization does the U.S. represent, anyhow? And is it the same civilization as the one making AI in San Francisco?

China’s government is above all focused on its own civilization (though its contours are unclear), and it aspires in a way others do not, to provide a universal answer to the material needs of its masses via total omniscience: Today, AI is a central plank in this plan to create a society where everybody is made visible by the cameras on every street.

The Adversary

“You and I meet on the dark sea of night; you have your direction, I have mine.”

—Xu Zhimo, in the poem, 偶然, or “By Chance”

Is the civilization that created ChatGPT called America? Is it called capitalism? Is it a repository of images, experiences and ways of being that is so vast that no one member of the civilization could know about it all? Whatever constitutes that civilization — old issues of The New York Times, Victorian novels, academic papers, Hollywood dialogue, Reddit posts — it built the West’s psychic abode and is now being chucked into the fire of its AI models.

Just as an unborn child is nourished by vitamins and nutrients from the mother, so an AI feeds on the complete sum of all our thoughts. Just as a child, once grown, experiences things unimaginable to their parents, so an AI model may think in ways that humans do not. But that does not change the fact that it is human progeny.

“With its sunny optimism and tendency to praise, ChatGPT is a Californian, one who has never suffered nor been held accountable for the consequences of its innovations.”

With its sunny optimism and tendency to praise, ChatGPT is a Californian, one who has never suffered nor been held accountable for the consequences of its innovations. The immense voltages of electricity it consumes, the lawsuits over which content can and cannot be used to train it: American AI is a massive temple in its own image, one which, once built, may transcend the limited imaginations of its creators.

In 1949, George Kennan, the architect of America’s Cold War strategy, wrote:

“The evils caused by a laissez-faire attitude toward technological advances have already produced illnesses which can only be cured by a high degree of paternalism. Only some form of a benevolent authoritarianism could manipulate living patterns in a manner adequate to restore a framework for healthy and vigorous citizenship.”

In his time, a paternalistic state emerged; today, the U.S. government intervenes in the lives of its citizens in new ways. Lurking like a shadow on the wall is China’s governance model, which its propagandists describe in ways that resemble Kennan’s ideal of benevolent paternalism.

If American AI reflects the country’s thoughts and cultural values, then it will inevitably reflect the disordered, chaotic mind of a society built around a burning desire for freedom — an irrational, often dysfunctional desire — to venture into the unknown, to search the skyline ahead to distract themselves from the mundane reality of the present.

Throughout the history of English-speaking peoples, many have sought to venture into the unknown, to master it, to conquer it. Why? Many of these explorers didn’t know; they just went. And so it is today. For Americans, going in search of the unknown requires no specific reason, no certainty of profit — although profiting thereafter is welcomed. The movement from a comfortable, familiar present into an exciting unknown is a justification of its own.

Many of the Chinese I have known find this confusing. What’s their real motive? The motive is life itself, which is a synonym for exploration in American culture. Not in China’s, though. Chinese AI seeks to make the wretched and the deprived into humans for the first time, raising them from the bestial conditions of poverty. The goal of the Chinese government and also, cultural customs, language and whatever else unites the population, is for the Chinese to live happily ever after, as a precondition for the stability of the apparatus itself.

Watching The Watchers

From the intertwinement of these dueling visions, a future artificial sentience will be born. But is it really possible for people to escape their own experience of reality and appreciate a different human culture on its own terms?

Some people call it “great power autism”: the tendency of a vast and diverse society to assume its norms are universal. Chinese policymakers often seem trapped in cycles of reaction to what they perceive as America, searching for a hidden logic in its actions, often assuming the worst — and also assuming a degree of central organization, of intentionality, that may not exist. In China, there is indeed a secret central committee that watches, decides and makes plans, so it’s easy enough to assume that America has one as well.

For their part, Americans — or at least some parts of American society — sometimes project ambitions on China that seem to emerge from guilt over their own repressed crimes: Chinese society is characterized by ethnic segregation; the Chinese aspire to world domination; they pretend they want to help poor countries, but actually they just want to do business there.

Because “China” is a social concept rather than a scientific one, it can be irresistible to use the metaphor of a single social body when discussing how its state plans mesh with funding structures, regulatory structures and population flows. That’s especially true given that the planning system and those regulating and funding AI in China all belong to the Chinese Communist Party. This degree of societal coherence (or totalitarianism) is unknown in the U.S., making it particularly difficult for Americans to identify with.

“In other words, America is a society where the brilliant thrive and the average suffer. China is precisely the reverse.”

As the AI boom matures, American commentators increasingly argue that, despite being deprived of certain cutting-edge chips, the Chinese may be more proficient at finding practical uses for the new technology. If the world of American capitalism has, in its late stage, tended to create infinite specialization and a society in which every person is constantly competing with every other, China’s patriarchal rule seeks to make a spot within the hierarchy for everybody, as long as they conform to it.

In other words, America is a society where the brilliant thrive and the average suffer. China is precisely the reverse. More complex societies are more costly to maintain than simple ones. Is the American social structure more fragile than the Chinese one? Does that mean that China’s approach, more prosaic than poetic, focused on meeting material needs rather than searching for the unknown and divine, will be more likely to succeed? It’s hard to imagine American AI succeeding right now, if only because there is no consensus on the definition of success. Chinese AI will be successful if the specific goals of the five-year plan are met.

Technology is designed to solve our problems, but in doing so, it engenders a new set of problems. As the bones of the world split open, shattered by powerful new analytic technologies, we can see the marrow that was always within. The prism is not the light, but the medium that makes light visible. So too with AI: It refracts our civilizational selves, making visible the invisible lives and instincts within the body of a society.