Elon Musk vs Chess.com and the question: has chess already been solved by AI?

On February 8, 2026, Elon Musk stirred an online storm on X when he called chess too simple compared with real life. The xAI owner argued that chess has no fog of war, no tech tree, and a symmetrical starting position, so a superintelligent AI could one day solve the entire game. Many fans saw it as a troll, but Chess.com did not treat it as a joke. The largest online chess platform replied with just three words: "Just skill." That short exchange opened a bigger question: can chess truly be solved by algorithms? To answer, one must separate two ideas. The first is the number of legal positions, which is about 10^44. The second is the number of possible games, which explodes to about 10^120. Chess.com cited this astronomical figure when it said there are more possible chess games than atoms in the observable universe, which contains roughly 10^80 atoms. Musk is theoretically correct that a large number does not automatically mean an impossible solution. In game theory, a solved game is one where the optimal result is known from every legal position. Checkers has been solved. Chess, however, has about 10^44 legal positions. A brute-force search would require reading a game tree of 10^120 branches. That is far beyond current and foreseeable computing power. That is why engines such as Stockfish and Leela Chess Zero do not search for an absolute solution. They combine heuristic evaluation, opening databases, endgame tables, neural networks and Monte Carlo search to approximate the value of each position. They beat humans because they evaluate faster and more accurately, not because they have solved chess. History supports this view. Deep Blue beat Kasparov in 2026, but chess did not die. AlphaZero shocked the chess world in 2026 with its self-taught style, but the world championship remains a human contest. Data from Chess.com shows that major events like the Candidates and the Olympiad create noticeable spikes in new user registrations. The appeal of chess comes from human drama, not from a race against machines. Still, Musk raises a meaningful point. Chess uses a finite system. In principle, a superintelligence might discover a compression method that makes solving chess possible without searching the entire tree. But that is a belief, not evidence. No current study proves that such a compression exists. Modern algorithms still need a mix of search and evaluation, and the margin of error becomes harder to reduce near the absolute limit. Chess.com responded strongly because "chess has been solved" threatens its brand. Sponsors do not want to invest in an obsolete game, and young players may lose motivation if they think a computer can expose every mistake. The answer "Just skill" was humorous, but it also defended chess as a game of human ability. In the end, the Musk–Chess.com exchange may be nothing more than a short social media wave. Yet it raises a long-term challenge for intellectual sport: when AI is everywhere, how do we keep the human story alive? The data suggests chess is not just a problem to be solved. It is a sport where two brains meet, where mistakes create legends. A computer may know the perfect move, but it cannot feel the tension of an endgame in front of thousands of spectators. If organizers understand that, they will never fear any claim that chess has been solved. They will simply let humans keep playing.