In 1997, Deep Blue defeated Garry Kasparov, and chess was supposed to die. The game that had defined human intellectual superiority for centuries had been conquered by a machine that could calculate 200 million positions per second. The obituaries were written. Human chess was declared obsolete.
Twenty-nine years later, professional chess is more popular than ever. Prize pools have grown. Viewership has exploded. Young grandmasters emerge from every continent. The game hasn’t just survived the age of superhuman engines—it has flourished in ways that would have seemed impossible in that moment of apparent defeat.
This is not merely an interesting fact about chess. It is perhaps the most important case study we have for understanding what comes next as artificial intelligence surpasses human capability in domain after domain.
The Paradox of Continued Play#
The first question worth asking is simple: why do people still play chess?
Engines don’t just beat humans—they humiliate us. Stockfish, the leading open-source engine, would crush Magnus Carlsen with the same ease Carlsen would crush a casual player. The gap is not close. It is not competitive. By pure performance metrics, human chess is objectively inferior in every measurable way.
Yet the stadiums fill. The streaming numbers grow. The sponsorships multiply.
The conventional explanation is that we watch chess for the drama, the narrative, the human element. We care about Carlsen’s psychology, about underdogs and rivalries, about the stories we can project onto the sixty-four squares. This is true, but it is incomplete.
Something deeper is happening. Humans continue to play chess because the struggle itself contains meaning that the outcome cannot erase. The process of a human mind grappling with complexity, making decisions under uncertainty, occasionally finding beauty in the fog of calculation—this has value independent of whether a machine could do it better.
This is a profound insight that cuts against the utilitarian logic that dominates our thinking about work and value. We have been trained to believe that the point of doing something is to produce an optimal outcome. If a machine produces better outcomes, human effort becomes worthless. Chess proves this logic wrong in a way we can all observe.
What Chess Got Right#
The chess world made several decisions, mostly by accident, that allowed human play to coexist with superior machines.
First, they embraced engines as tools rather than fighting them as threats. Today’s grandmasters train extensively with engines. They use them to prepare openings, analyze games, discover new ideas. The human-engine collaboration has pushed human understanding of the game to heights that would have been impossible without machines. Players didn’t resist the technology—they absorbed it.
Second, they maintained clear boundaries during competition. You cannot use an engine during a rated game. Elaborate anti-cheating measures ensure this boundary holds. The playing field is explicitly human, even as the training field is hybrid.
Third, and perhaps most importantly, the chess community never abandoned the belief that human chess has intrinsic value. They didn’t apologize for their limitations. They didn’t argue that human chess was somehow “better” than engine chess. They simply asserted, correctly, that human chess is a different thing—and that this different thing matters.
The Failure Modes We Must Avoid#
Not every domain has navigated this transition well. And as AI extends into areas far more consequential than board games, the failure modes become increasingly dangerous.
One failure is denial. Pretending that human capability isn’t being surpassed, clinging to the belief that machines will never truly match human judgment, creativity, or understanding. This is the path of irrelevance. Chess players who refused to engage with engines in the late 1990s quickly found themselves outclassed by those who did.
Another failure is capitulation. Concluding that if machines are better, humans should simply stop trying. This leads to atrophy—of skills, of judgment, of the very capacities that make us capable of directing machines wisely. A world where humans defer entirely to AI recommendations is a world where humans lose the ability to evaluate those recommendations.
The third failure is the one that should concern us most: chaotic integration. Not thoughtful merging of human and machine capability, but haphazard adoption driven by whoever moves fastest, whatever feels easiest, whatever serves short-term incentives. This is where the nefs—the ego, the impulses, the animal drives that override stated values—take over.
Chess avoided this because the stakes were low enough that no one was in a hurry. AI in medicine, finance, governance, and war does not have this luxury.
The Merge Is Coming#
There is no future where humans remain separate from AI. This is not speculation. It is already happening. The question is not whether we merge, but how.
The chess model suggests one path: humans enhanced by AI tools, but with clear domains where human judgment remains primary. Grandmasters use engines to train, but they play as humans. Doctors might use AI to diagnose, but they treat as humans. The hybrid is more capable than either alone, and the boundaries preserve something essential about human agency.
But this path requires intentionality. It requires deciding, collectively, where the boundaries should be. It requires resisting the pressure to automate everything that can be automated, simply because it can be.
And here is where the analogy breaks down, because chess is just a game.
When we talk about AI in work, in governance, in the shaping of minds through algorithmic platforms, the stakes are not a tournament trophy. They are the structure of society itself. Who decides where human judgment remains essential? Who enforces the boundaries? Who benefits from the erosion of those boundaries?
The answers to these questions will not emerge from market forces optimizing for efficiency. They will emerge from power—who has it, who wields it, what values guide them.
The Sustainability Question#
The original tweet framed this as a question about sustainability. How do we use AI in a sustainable way?
Chess provides half an answer. Humans kept playing because they found meaning in the struggle, not just the outcome. They used engines to become better, not to become obsolete. They maintained spaces for purely human expression.
But sustainability requires more than individual choices. It requires systems that don’t collapse under the pressure of optimization.
If AI-generated art can be produced infinitely at near-zero cost, what sustains human artists? If AI can write code faster and cheaper than any developer, what sustains the human understanding necessary to direct that code wisely? If AI can optimize political messaging beyond any human’s ability to resist, what sustains democratic deliberation?
Chess is sustainable because no one’s livelihood depends on being better than Stockfish. The same cannot be said for most of human labor.
This is the harder problem. And the chess analogy, while illuminating, cannot solve it.
What We Must Choose#
The future is not determined by technology. It is determined by decisions—made by those with power, shaped by those with voice, endured by everyone else.
The chess paradox shows that human endeavor can retain meaning even when machines surpass us. But it does not guarantee that meaning will be preserved. It does not guarantee that the boundaries will hold. It does not guarantee that the merge will be thoughtful rather than chaotic.
What it shows is that the outcome depends on what we choose to value. If we value only outcomes, only efficiency, only optimization, then human effort becomes noise to be eliminated. If we value the process, the struggle, the irreducible something that humans bring to any endeavor, then space remains for us.
The real question—the one that chess cannot answer—is whether we are capable of choosing wisely. Whether our nefs will drive us toward whatever extreme feels easiest: complete resistance or complete surrender. Whether we can find the difficult middle path that preserves what matters.
History suggests skepticism. Human nature does not change, even as circumstances transform beyond recognition. Every technology is captured by the same drives that captured every technology before it.
But history also shows that sometimes, in some domains, for some time, humans get it right. Chess got it right. The question is whether we can replicate that success in the places where it actually matters.
Uncertain.