- OPENAI O3 defeated Elon Musk’s Grok 4 in chess
- Magnus Carlsen delivered comments on the quality of Grok’s logic
- Grok 4 made repeated mistakes, while O3 played stable
The AI chess tournament between OPENAI’s O3 model and Grok 4 of XAI invited many speculations such as a kind of proxy battle between the two companies and their respective CEO. However, any comparison with the days of Deep Blue and Bobby Fischer soon vanished, since Openai O3 repeatedly annihilated Grok 4, winning four games in a row, accompanied by the mocking comment of the former world champion of Chess Magnus Carlsen and the great teacher David Howell.
The confrontation occurred in Kaggy’s Game Arena, a digital coliseum where the AI models the models in chess and other games. The tournament had eight of the most prominent LLMs in the business: O3 and O4-Mini de OpenAi, Gemini 2.5 Pro and Flash of Google, Claude Opus de Anthrope, Deepseek and Kimi de Moonshot, and Grok 4 of Xai. The final was reduced to Grok and O3, but Grok’s performance in the last round did not seem like a battle of champions.
Carlsen and Howell deviated between serious comments and roasted when Grok’s performance seemed erratic. In the first game, he quickly sacrificed his bishop, then began to exchange pieces as if he were hurried to go home. Things did not improve in the next Grok game.
“[Grok] It’s like that guy in a club tournament that has learned the theory and literally knows anything else, “Carlsen said during the second game.” Makes the worst mistakes after that. “
Grok’s yield was so out of the fictions that Carlsen described him around 800 Elo, or slightly above a beginner. He gave O3 a modest but respectable 1200, in the middle of most hobbies players. Although O3 did not play brilliantly, I didn’t have to do it. He played solid chess. He didn’t go crazy. He turned its advantages and carried out the classic chess movements.
“The O3 is quite ruthless in conversions; it looks like a chess player. It seems that Grok learned some opening movements and knows the rules, but not much more,” said Carlsen. “Grok’s movements are movements related to chess. They simply arrived at the wrong time and in strange sequences.”
Chess AI
Chess was not the main point of the tournament, despite its prominence. It was about how general use models handle events with strict rules such as chess games. It turns out that they are not great, but O3 is the best of the limited sample. As AI is embedded in everything, the ability to follow the rules and points patterns becomes essential. Chess is a transparent unique way of observing that. Or you did the correct movement or did not. When a model plays well, you can see logic; Otherwise, the queens fall as dominoes, and the game becomes as confused as that metaphor.
Chess is a window of how well an AI can plan, evaluate options, avoid catastrophic errors and stay logically consistent. If Grok throws a queen because he does not understand long -term consequences, what could he do in a legal document or when reserving trips?
That the final was between Openai and XAI added a drama with Sam Altman and Elon Musk in Loggerheads in public. The chess final did not solve the battle between them, but gave OpenNAi a victory in public relations in the field of public perception, and a limited but very real compliment of Magnus Carlsen.
