Computer or person

MG Emil Sutovsky recently published shocking news on his Facebook page: the computer program “Ponanza” beat the best shogi player in the world. But that is not all! The program applied the novelty on the first move: no professional shogi player had ever played like this before! I think this is about how to start a debut with 1.Ka3 or 1.f3.

Of course, this is not so shocking news for chess players. Unfortunately, we have already left this stage of computer development far behind.

Today, the top-end program can easily beat Magnus Carlsen by playing literally any initial move. The future in human and computer competition is pretty obvious: every day the gap between us and the machines is growing.

The question is different: how will the irreversible progress of chess programs affect the human game?

As Yogi Berra correctly said, it is difficult to predict anything, especially regarding the future. Indeed, no one knows how tournament chess will look in 10 years. I would like to discuss only the obvious trends that we observe today.

The decisive moment in the struggle between man and the computer was 1997, when the program “Deep Blue” beat the world champion Garry Kasparov. Then even the most optimistic proponents of the theory of the superiority of the human mind over the computer realized that we could not win in such a competition. Then instead of the mantra “a person is smarter than a computer” we began to repeat “yes, a computer is stronger than a person, but he does not understand chess, but only considers it better!”

In fact, there are many parties supporting this claim. Take for example the three moves that shocked the world during the first installment of Kasparov’s match against Deep Blue:
Three completely unrelated moves that make Black’s position more vulnerable looked like a beginner’s game. And yet they are played by an entity that has managed to beat the world champion!

And now back to our time. The computer game has become more like a human one (or maybe we began to play like computers?). Nowadays, the idea that a person understands chess more than a machine has quietly disappeared into the air. In truth, sometimes it seems to me that computers understand much more than us in a positional game.

A striking example is the recent game played in a super tournament:
When I watched this game, I expected that the rivals would agree to a draw at any minute, especially when both rooks were exchanged and the following position arose:

Now see how the game ended:
While most of the players would really agree to a draw at the very moment when the queens were exchanged, the computer continued to show White’s advantage in half-pawns throughout the endgame! Now tell me, who understands chess more?

If we do not mention the problem of computer cheaters, then the biggest influence on chess was made by computers in terms of openings. An abundance of parties with Berlin defense in top tournaments is no coincidence. It is much easier to analyze positions with a smaller set of figures, and therefore the Berlin endgame has become a favorite tool of most leading grandmasters.

Chess players are digging deeper and deeper with the help of their silicone friends, in the end trying to solve a chess game. Some of them get very close! In a recent MG interview, Boris Avrukh mentioned one episode from his practice of working with MG Wesley So’s second world number. Wesley showed the option for Black, which after 50 moves forcedly leads to a rare ending with two knights against the rook and bishop. When Avruh asked if So knew how to draw, Wesley instantly showed how to build an impenetrable fortress!

And even if it’s a really very impressive episode, it’s also very sad.

At some point, it will be necessary to do something in order to return the human element to the opening game. I don’t think that Fischer chess (Chess960) is a solution, since the starting position is too far from classical chess. Just imagine: we teach our children the game of chess using the beautiful games of Morphy, Fisher and Kasparov as an example, and then we suddenly say: sorry, children, but now you play Fisher’s chess and figure it out yourself!

In fact, computers can help us solve this problem. If you ask the computer engine to evaluate the starting position, this will give you about a quarter pawn advantage for White. But if you force White to play 1.h3, then the score drops to an almost dead draw! Something like 1.a3 e6, according to the computer, also leads to an absolutely equal position. So why should we turn to Fischer chess if we can find a dozen opening positions in classical chess, which the computer evaluates as equal, and the theory is completely absent from them?

The current world champion successfully demonstrates how to play classical chess without a debut theory:
Another trend that I foresee is dramatic changes in the way I write debut monographs.