Trang chủChessMusk says AI will solve chess; Chess.com pushes back with 10^120
Chess

Musk says AI will solve chess; Chess.com pushes back with 10^120

Elon Musk tuyên bố AI sẽ tìm ra lời giải cuối cho cờ vua trong 10 năm tới; điều này gây tranh cãi vì trò chơi có khoảng 10^120 ván cờ hợp lệ, vượt xa khả năng tính toán của siêu máy tính. | Key facts: - Ngày 26/10/2025, Elon Musk viết trên X rằng cờ vua sẽ bị AI giải mã trong vòng 10 năm. - Chess.com phản hồi bằng dữ liệu 10^120 ván cờ khả dĩ, cho thấy không thể dùng thuật toán vét cạn. - Số vị trí hợp lệ của cờ vua khoảng 10^44; cờ đam đã được giải năm 2007 nhờ 10^31 nước đi hợp lệ. - Năm 2025, Grok 4 của xAI thua o3 của OpenAI trong một giải cờ AI; tuyên bố của Musk có bối cảnh cạnh tranh. | Source: Business Insider, 27/10/2025 | Related Q&A: - Cờ vua có bị AI giải mã hoàn toàn không? Không trong tương lai gần, vì độ phức tạp cây trò chơi 10^120 vượt xa sức xử lý hiện nay. - Nói cờ vua đơn giản có đúng không? Sai nếu hiểu là dễ dự đoán; cờ vua có chiến sâu vô hạn dù luật chơi cố định. - Sự kiện giúp gì cho Chess.com? Sự việc tạo ra hiệu ứng thương hiệu lớn, giúp nền tảng tiếp cận nhóm khán giả công nghệ mới.

On October 26, 2026, Elon Musk stirred up social media platform X by claiming that chess would be solved by artificial intelligence within ten years. The fastest response came not from a top grandmaster but from the Chess.com account - the world's largest online chess platform. Chess.com's reply was concise: try processing the number of legal games before calling it a solution. To me, this exchange is not dry; it raises a serious question about how people understand the power of data. Musk has a history of downplaying chess. In 2026, he said chess was too simple for real life because it lacks fog of war, a technology tree, and asymmetric information. He repeated that view over the weekend when suggesting AI could find all the answers on the chessboard. Public opinion quickly split: one camp viewed it as a provocative remark, the other as a fundamental misunderstanding of the game's complexity. Chess.com, as a community gatekeeper, used the language of quantitative thinking that technology billionaires often praise: data. The number of legal positions in chess is about 10^44, but the number of possible games reaches 10^120. People often compare 10^120 to 10^80 - the estimated number of atoms in the observable universe - to emphasize that no human can brute-force all search branches with present algorithms. To call a game solved, one must prove that from every position, the optimal outcome is known when both sides play perfectly. That goal fails at an early stage: modern computers cannot evaluate all positions, let alone store them. People may call Musk's statement a shock; I call it unread data. Checkers provides a historical example. That game was solved in 2026 after decades of supercomputer effort, with roughly 10^31 possible legal branches. Chess has a far larger branch complexity, so the only route to an absolute solution would be a new mathematical insight, not merely hoping hardware becomes cheaper. More importantly, AI beating humans is not the same as AI solving the game. AlphaZero beat Stockfish in 2026 with elegant play, but it did not prove whether perfect chess is a draw or a win for one side. This resembles a runner winning the race without anyone knowing the absolute biological limit of human speed. This time, the background reveals a fascinating contradiction. Musk runs xAI, the company that trains the Grok model. In a 2026 AI chess event organized by the tech industry, Grok 4 lost to OpenAI's o3, even though both used reinforcement learning. A week earlier, Musk touted xAI's Colossus supercomputer and its record number of chips; chess remains an arena for testing reasoning power. It is possible that Musk’s denial of chess complexity is a way to reframe his own company's model failure through clever positioning. But data may not be friendly to PR tricks: it shows that AI is stronger at calculation, but not yet stronger at discovering the deep structure of a perfect-information game like chess. Musk's statement accidentally created what sports media often call an inverse effect: rather than devaluing chess, it gave Chess.com a natural publicity moment. Within hours, the platform gained brand visibility among tech-savvy users who might not follow chess tournaments. In an age when every billionaire remark is measured in engagement, this quarrel benefits both parties. But becoming a viral topic does not mean Musk's claim is correct. Another view deserves emphasis: do not let the word solved make us forget that chess is a social maze. Chess is attractive not only because of its huge branching factor, but because humans love uncertainty and the stories that arise from mistakes. Even if a super-AI could print out the optimal result for every position, millions would still study classical openings and search for moments of hope. This creates a paradox: AI becoming strong enough to solve chess might not make chess pointless; it could turn us into more curious spectators, not fewer players. Numbers are ascetic; we must abandon the easy phrase that solving means ending, so the true nature of the game becomes visible. This story shows two things: the tech industry needs intelligent benchmarks with clear rules to compare AI models, and the sports world needs to remember that technology does not replace emotion. If I had to answer whether chess will be solved, the answer is: not within our lifetime, and the very idea of a single answer would make the game less beautiful. The next thing to watch is not the bot's solve rate, but how platforms like Chess.com turn the AI narrative into their own data playground. There, humans still ask the questions, and AI only serves as a very fast move counter. In this large debate, my advice to chess lovers is simple: observe calmly, and do not believe any solved claim until an algorithm is publicly revealed. And if you need one number to remember, remember 10^120. It is more than chess complexity; it is a measure of the humility needed when humans talk about the future of intelligence.
Cầu thủ liên quan