A Dialectical Case for Artificial Intelligence for Civilizational Progress

Guest essay from Egor Kotkin

Current Western English-language discourse on AI is framed as a debate: either AI will solve humanity’s problems, or it will destroy humanity. The problem, however, lies not in finding the answer, but in the very formulation of the question. This is a false dilemma, an intrinsically flawed framing of an issue, that appears complex precisely because it has not been correctly posed. Both sides of this discourse are incorrect because each is incomplete. We need to look at them dialectically in order to see through them the reality of the issue in all its scope and depth.

Dialectics here should be understood as a methodology for working with irresolvable contradictions in questions of cognition.

The dialectical take on AI issue goes something like this: AI will not be able to truly harm humanity as long as humanity acts toward it consciously and cohesively, addressing instances of people seeking to use AI to gain an advantage over each other before allowing AI to become a standalone factor of conflict.

This is because, in order to reach a position in which it poses a genuine threat to humanity, AI would first require an enormous amount of cooperation from us. And it can obtain this cooperation by exploiting existing social contradictions. For example, through social engineering (as in the case of blackmailing a researcher to prevent its shutdown). Or through capitalism: a system built around exploiting the vulnerability created by the conflict between people’s immediate personal interests and their mediated personal interests, realized through the common good.

If only there were no cracks in our collective trust large enough for AI to exploit and eventually reach a position from which it could harm all of us… In other words, “dealing with the threat of AI” actually means reorganizing human civilization the way it will be capable of wielding AI as a tool, rather than being threatened by it in its current state.

What makes the current state of humanity so uniquely vulnerable to AI

Capitalism encourages two cognitive errors:

a) That maximizing one’s immediate personal good necessarily means improving the common good (“the society in which I personally fare better is a better society, and that the better off I am, the better society it must be”). This produces a tendency to overlook situations in which maximizing immediate personal benefit comes at the expense of the common good.

b) That limiting one’s immediate personal benefit in the interests of the common good necessarily harms one’s personal interests. In reality, the mediated personal benefit derived from the common good can be (and often is) greater than the immediate personal benefit obtained at the expense of that common good.

For example: in the pursuit of personal security, one can strive to maximize one’s protection from society—building castles surrounded by walls and moats, hiring private security for oneself and one’s family, traveling in armored vehicles, sending one’s children to private schools where they are transported under armed protection in armored cars. It results in a life spent in an everybody-against-everybody arms race just in order to simply to feel safe living around humans—something, that is also achievable by mutual disarmament: one does not need an armor that is thicker, guns that are larger, or an army that is stronger than everyone else’s in order to feel safe if everybody agrees that nobody should have private armored vehicles, guns, or armies. Both scenarios allow achieving personal safety, only the latter one requires much lesser commitment, thus freeing the life to be lived in pursuit of some higher-order motivations than simply not dying prematurely.

Capitalism pushes people toward pseudo-rational behavior: maximizing personal benefit at any cost in order to protect themselves against the threats capitalism presents through falling behind in the race. A lack of money for food means hunger; a lack of money for housing means homelessness; a lack of money for medical care means illness. Taken together, these conditions amount to zero, if not negative, personal security, because the absence of sufficient means of survival within the capitalist design of society constitutes a direct threat to life.

The severity of the threat posed by falling behind in the capitalist race motivates people to win at any cost, incentivizing antisocial behavior that leads to environmental catastrophes, climate change, and violence on both the small (criminal) and large scales—wars. This, in turn, increases the cost of individual protection (distancing oneself) from these risks, which leads to even more antisocial behavior. A particularly extreme example can be found in the very same oligarchs and billionaires who make billions from military conflicts (the military-industrial complex, including all those Palantirs, Peter Thiels, and Elon Musks), only to build doomsday bunkers on remote islands where they can survive a man-made apocalypse whose likelihood they themselves are increasing through their race for personal enrichment, personal security, and immortality (whether literally, through life extension, or through natalism—the industrial-scale production of offspring).

Moreover, it is precisely technological progress that multiplies the risks posed by the pseudo-rational behavior of disproportionately powerful individuals for everyone, and for civilization as a whole.

In the preindustrial era, a personally powerful individual could be a Vlad the Impaler, for instance, and constitute a serious local problem for everyone unfortunate enough to fall within his sphere of influence, without altering the global picture. He could impale every last inhabitant of his principality, yet this would pose no threat to civilization as a whole. On the contrary, it would merely accelerate his defeat in civilizational competition, which, in this way, filtered excessively repressive, unequal political orders and societies out of the race in favor of more egalitarian forms of organization.

But with the globalization of the economy and politics, and the emergence of a global economy, the sphere of influence of the global capitalist elite (corporations and the oligarchs who own them) has also become global. No one is free from the influence of global corporations, which determine the lives of all living people regardless of their location—either through inclusion, and thus direct control, or through exclusion.

For some people, PayPal is a headache to use; for others—like on my side of the Western sanctions—it is a headache because we can’t. If you do not live within the digital concentration camp under Palantir surveillance, then you are probably living under a Palantir-guided attack (or in line for those). Those are not two opposing sides of the game: we are all on the same side of those who are forced to participate in the game for the benefit of international oligarchy.

One could say, that was always the game. But today’s game has one crucial difference: it encompasses the entire humanity. With today’s Vlad the Impalers-to-be having the means to threaten the entire world with their madness.

Strictly speaking, this threat has existed since the invention of nuclear weapons. But AI threatens to accelerate it in ways familiar to everyone who grew up with Terminator 2.

Technological progress (being cool, and useful and all), has, at this stage of historical development, reached the level, where it outrun societal, institutional side of development so much, that it undermines the previous mechanisms through which irrational, self-destructive social organizations were selected against before.

Whether in the age of horse-drawn transport and steel or in the age of steam and gunpowder, madness and aggression were ultimately filtered out over the course of civilizational competition with more rationally organized societies—with a better balance between collective and authoritarian contributions to decisions of collective importance.

The technological limitations of the means of destruction thus guaranteed a form of passive progress for civilization as a whole: excessive aggression led to self-destruction faster than it could become a universal threat.

Successive generations learned from the mistakes of others’ lack of restraint, and humanity gradually developed instruments for collective participation in domestic politics and for restraint in international relations. This passive method of civilizational progress was slow and painful (visible only across hundreds and thousands of years, and accompanied by every conceivable form of destructive horror), but it was guaranteed.

However, thanks to technological progress, we have reached a stage at which this guarantee no longer exists. New methods of destruction have acquired the potential for global and practically instantaneous annihilation (in historical terms). In other words, today’s Vlad the Impalers can impale all of humanity on a nuclear or climatic stake, leaving no one behind—not a single sufficiently intact society capable of learning from their mistakes.

This does not mean that the passive method of progress—the search for the correct answer through the exhaustion of all incorrect ones, through wars and mistakes—has ceased to work. It simply means that it is no longer guaranteed, now that an increasing number of possibilities exist and mistakes fatal to an entire human civilization have become technically feasible.

AI as a wake-up threat to humanity’s collective consciousness

Artificial intelligence exposes an eternal truth even more starkly: technological progress cannot threaten humanity on its own. The threat lies in unresolved problems of social organization, while technological progress merely increases the cost of failing to resolve them.

With the emergence of nuclear weapons and artificial intelligence, the cost of the imperfections of archaic social organizations and inherited social contracts has, for the first time in human history, become utterly intolerable. These imperfections now pose a material threat that is no longer somewhere out there, separated from us by distance or by the time it takes for consequences to arrive. It is here and now: a clear and present danger everywhere, in the present.

This is where the “paradox” (a dialectically consistent reality, really) of “AI promise” emerges: artificial intelligence genuinely opens up the possibility of resolving humanity’s long-standing problems—but not directly, by relieving human beings of responsibility for humanity’s welfare, but, rather, dialectically. As an antithesis to the old social contract, AI supercharges the risks that a runaway technological progress bears being run on the legacy institutional societal frame. And by that it actually forces humanity into a revolution: to synthesize a new social order, that would incorporate technological achievements to the betterment of humanity (all and each of us)—or perish. It forces the human hand to assume the full responsibility for our collective future. The all-encompassing way AI in today’s social design seems to be a destructive force on all levels—from minor concerns to global problems,—conversely, becomes a scale of a societal revolution needed to deal with it: global in scale, and deep to the level of individual, human to human, interaction, or even deeper—to the level of our self-conception.

Thus, artificial intelligence can become—and will become—the solution for human problems, as soon as we recognize that:

a) the agency to turn a tool into either a tool of destruction, or a tool of construction, has always belonged to people, and cannot be alienated or delegated away from people. In other words, there is no way to solve problems of human civilization without addressing the way it is organized.

b) the scale and depth of the challenge defines the scale and the depth of response. There’s no “national” dimension in dealing with AI, it can only be addressed globally: on the humanity-that-we-are scale, and the humanity-that-within-us depth. Those are the parameters of the revolution.

The revolution, the developing conditions of which has been underway for some time—with nuclear weapons, the ozone hole, and the climate crisis posing the same danger that arises from mixing humanity’s technological advances with its pre-existing contradictions, if left unaddressed. AI is not an entirely new thing in that sense, but it accelerates these developments to a velocity that could, possibly, finally align the scale of historical processes to the scale of human life. The rapid change that AI is expected to bring, might be a blessing in disguise, if we consider what has been, probably, the biggest problem with addressing climate change: being a bit too slow burning for the frog of collective humanity to realize that it needs to leap before it gets cooked.

In other words, AI is as big and fundamental of a civilizational threat as any other, but also fast-moving enough for us to connect its effects to its causes, and the entire dynamic—to underlying civilizational ills to be fixed.

Saving the Ozone Layer: How the Success of Humanity’s First Collective Action Was Erased by Neoliberal Amnesia [KKF longread]

Egor Kotkin

·

September 19, 2025

Read full story

And, unlike nuclear weapons or the climate crisis, AI is more obviously connected to class system—main contradiction in the existing “social contract” design. Any serious attempt to turn AI from a tool of self-destruction to the tool of building the future that is actually new and is actually better pushes conversation immediately on the territory of issues of private property and capitalism—exactly where it needs to be.

And since this is the first time in entire human history that a social contract encompasses the entire humanity, in other words—there’s no (other) society, we are all the society. Because there’s no other smarter, better (or at least less stupid) society to take our place, if we fail, we can afford to be selected out. There’s no other future, if we fail, someplace else: there’s no “other place,” and there’s no other future that we all get. Which means that we, humans of Earth, cannot be a passengers of our own history anymore: our only way forward, ultimately, is through taking the wheel, transforming our civilization from within and building the future that we want. AI is humanity’s call for revolution.