Can AI Ever Understand the Human Soul? Revisiting Notes from Underground in the Age of Algorithms

Is AI just another attempt to rationalize the irrational? In Notes from Underground, Fyodor Dostoevsky presents a narrator who rejects the Enlightenment ideals of reason and logic, embracing instead the chaotic, unpredictable nature of human existence. In an age where artificial intelligence is increasingly used to predict and control human behavior, Dostoevsky’s cynical critique of rationalism feels more relevant than ever. Can an algorithm ever truly comprehend the irrational depths of the human soul?

The underground man, Dostoevsky’s unnamed anti-hero, is a deeply contradictory character. He both embraces and despises his own suffering, all while railing against the idea that human beings can be neatly categorized or reduced to rational actors. "Man is sometimes extraordinarily, passionately in love with suffering," he muses, rejecting the notion that reason alone can explain human desires. This philosophical rebellion could serve as a cautionary tale for modern AI, which thrives on rationalizing and predicting human behavior based on patterns and data. But what happens when those predictions clash with the innate messiness of human emotions, desires, and contradictions?

The Limits of AI Rationalism

Much like the rationalists of Dostoevsky’s time, AI systems are built on the premise that with enough data and sophisticated algorithms, human behavior can be predicted and, in some cases, controlled. AI thrives on patterns—on reducing human actions to predictable models that can be monetized, streamlined, or influenced. This belief in the power of logic, data, and reason to solve all human problems is reminiscent of the very rationalism that Dostoevsky despised. The underground man’s refusal to conform to society’s expectations of logical behavior is a reminder that humans are not, and perhaps never will be, predictable data points.

Consider, for example, how AI is currently used in the criminal justice system. Algorithms like COMPAS are employed to predict the likelihood of recidivism based on a defendant’s past behavior, demographics, and other factors. These models claim to offer rational, data-driven insights into human behavior. Yet they often miss the irrational, emotional, and chaotic elements that define so much of what it means to be human. As Dostoevsky’s underground man laments, "What does reason know? Reason only knows what it has succeeded in learning." AI, in a sense, suffers from the same limitation—it only knows what it can quantify, while human existence often defies quantification.

Rebellion Against Predictability

In Notes from Underground, the underground man rebels against the notion that human beings can be reduced to their base instincts or desires. He views himself as a "spiteful" man, one who revels in his contradictions and refuses to behave as expected. He mocks the idea of a rational "crystal palace" where human desires are perfectly satisfied, and everyone lives in harmony. For him, such a world would be intolerable precisely because it would erase the messiness, the suffering, and the unpredictability that make life meaningful.

Today’s AI-driven world often promises a version of the "crystal palace"—a future where algorithms optimize every aspect of our lives, from healthcare to shopping to relationships. But is this really what we want? Dostoevsky’s underground man would argue that there’s something deeply human about resisting the smooth predictability of a perfectly rational world. We don’t always act in our best interests. We make choices that are irrational, self-destructive, or emotionally driven—and that’s part of what makes us human.

AI and the Denial of Free Will

One of the most striking aspects of Notes from Underground is its critique of the denial of free will in the name of rationality. The underground man constantly asserts his freedom to act irrationally, even when it’s against his own best interests. He rejects the idea that humans should always act in accordance with reason or logic. "I am a sick man… I am a spiteful man," he declares, embracing the contradictions within himself.

AI systems, on the other hand, are designed to predict and, in some ways, limit human choices. When algorithms determine the content we see online, the ads we’re exposed to, and even the decisions made in courtrooms, they shape our behavior in ways that can undermine our sense of autonomy. Much like the underground man, we might find ourselves resisting these subtle forces of control—not because it’s rational, but because the very act of rebellion affirms our humanity.

Conclusion: AI Cannot Grasp the Irrational Depths

Dostoevsky’s Notes from Underground offers a powerful critique of the rationalist belief in the predictability of human behavior—one that resonates in today’s AI-driven world. While algorithms can analyze patterns, optimize decisions, and make predictions, they struggle to capture the irrational, contradictory nature of human existence. The underground man’s refusal to be reduced to a rational model of behavior is a reminder that, in the age of AI, we must guard against the tendency to see humans as mere data points. The richness of our experience lies not in our predictability, but in our capacity to defy it.

References:

Dostoevsky, Fyodor. Notes from Underground. Vintage Classics, 2009.

Zuboff, Shoshana. The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. PublicAffairs, 2019.

Harari, Yuval Noah. Homo Deus: A Brief History of Tomorrow. HarperCollins, 2017.

Eubanks, Virginia. Automating Inequality: How High-Tech Tools Profile, Police, and Punish the Poor. St. Martin’s Press, 2018.