AI’s superiority does not necessarily make human participation obsolete.
We often assume that once machines become better than humans at an activity, humans will naturally stop doing it. But the existence of a superior performer does not automatically destroy the activity. It can simply change why humans do it.
Chess is a perfect example. Modern chess engines such as Stockfish are vastly beyond human chess-playing ability. Yet people continue to play chess—not because they expect to outperform Stockfish, but because playing chess is itself valuable.
“The ability to play chess is the sign of a gentleman, and the ability to play chess well is the sign of a wasted life.”
— Paul Morphy
The irony is striking. A chess legend could regard extraordinary investment in chess as a “wasted life.” Yet humanity has since built machines that devote enormous computational resources to becoming extraordinarily good at the same game. And humans still play.
Stockfish also illustrates another phenomenon: the closer a system gets to the frontier of performance, the harder further improvement becomes. Obvious errors are eliminated first. What remains are tiny differences between already excellent choices.
Some classes of chess positions can now be completely solved through endgame tablebases, while the complete game remains vastly beyond exhaustive solution. At extreme levels, improvement becomes increasingly complex and expensive, while each additional increment may have little practical significance on a human timescale.
This pattern could eventually appear across professional cognitive skills. Coding, writing, research, mathematics and design are currently valuable partly because humans can perform them economically. But if AI becomes dramatically better and cheaper, the economic reason for humans to perform these tasks may weaken or disappear.
That does not necessarily mean humans will stop doing them. The economic usefulness of a skill and the human value of practicing it are two different things.
Adults are physically superior to children, yet children’s sports do not disappear. Professional athletes are vastly better than ordinary participants, yet ordinary people continue to play.
Nobody asks a child to justify playing football by demonstrating that they could defeat a professional. The activity has value beyond the performance hierarchy.
The same could happen with intellectual skills. Coding, writing and problem-solving could gradually move along a different spectrum:
necessity → profession → competition → recreation → craft
People might write without AI, code under constraints or solve problems manually even when AI can produce a better answer. The purpose would no longer be maximum efficiency. It would be the challenge, the learning and the experience of doing it.
“If you know exactly what you are going to do, what is the point of doing it?”
— Pablo Picasso
The question becomes especially relevant when AI can increasingly produce the destination before humans have begun the journey.
This leads to an important distinction between performance and participation. A machine can make a better chess move without making human chess meaningless. An AI can write better code without making programming meaningless.
“To travel hopefully is a better thing than to arrive, and the true success is to labour.”
— Robert Louis Stevenson
The value of human activity may lie not only in what is produced, but in what happens to the person while producing it.
This could lead to something like Professional Skills Olympics.
Humans could compete in coding, writing, engineering, mathematics, research, design and other professional skills—even when AI is objectively better at them.
AI would not necessarily be excluded. It could become the judge, coach, critic and benchmark, evaluating human performance against increasingly sophisticated standards.
The competition would no longer be about proving that humans are better than machines. It would be about testing what humans can achieve when being the best is no longer the point.
This also changes how we think about learning. Children learn mathematics even though calculators are better at arithmetic. People play music even though recordings can be technically superior. People paint even though cameras reproduce reality more accurately.
Learning does not have to exist solely to produce the world’s best output. It can exist to develop the person.
If automation removes the economic necessity of many professional skills, should society simply allow those skills to disappear? Or should we deliberately preserve opportunities for humans to practice, compete and develop them?
Government could support public competitions, educational programs and institutions for human skills—much as societies already support sport, arts and cultural activities that cannot be justified purely through economic productivity.
The goal would not be to pretend that humans remain superior. It would be to preserve human capability even when it is no longer economically optimal.
But recognizing the value of human participation should not become an argument against developing AI.
The fact that humans continue playing chess does not mean we should stop improving Stockfish. Humanity has always pursued difficult frontiers whose practical value was not immediately obvious.
“Why climb Mount Everest?”
“Because it’s there.”
— George Mallory
The same principle can apply to AI. We can continue pushing machines toward capabilities far beyond human performance without requiring those capabilities to make humans obsolete.
AI can keep climbing its Everest while humans keep playing their games.
The machine can optimize the frontier. Humans can explore the experience. AI does not have to remain inferior for humans to have a meaningful role, and humans do not have to remain superior for AI development to remain worthwhile.
We should build a human-centric society, with AI at the peripheral boundary of human life—not a machine-centric society with humans at its periphery.
Perhaps the future question will therefore not be: “What can humans do better than AI?”
That is an increasingly narrow foundation on which to build the meaning of human activity.
The more interesting question is: “What will humans continue to do even after AI can do it better?”
Professional Skills Olympics is one possible answer—a society in which AI becomes the benchmark, humans remain the participants, and mastery remains valuable even after superiority has disappeared.
The ultimate lesson of Stockfish may therefore be unexpectedly human.
We do not play chess because humans are the best chess-playing entities on Earth.
We play because playing chess is worth doing.
And we do not have to stop building better machines simply because humans cannot beat them.
AI may win the performance contest without winning the meaning contest.
The human game does not have to end when the machine becomes better at playing it.