- Openais O3 defeated Elon Musk’s Grok 4 by chess
- Magnus Carlsen delivered biting comments on the quality of Grok’s logic
- Fall 4 made repeated tabs while O3 played steady
The AI Chess Tournament between Openais O3 model and Xai’s Grok 4 invited lots of speculation as a kind of proxy match between the two companies and their respective CEOs. Any comparison with the days of Deep Blue and Bobby Fischer soon disappeared when Openai O3 repeatedly wiped out to the Four and won four games in a row, accompanied by the delineational comment from former World Chess Master Magnus Carlsen and Grandmaster David Howell.
Showdown happened at Kaggles Game Arena, a digital coliseum where AI models are fighting in check and other games. The tournament contained eight of the most prominent LLMs in the industry: Openais O3 and O4-MINI, Google’s Gemini 2.5 Pro and Flash, Anthropic’s Claude Opus, Moonshot’s Deepseek and Kimi, and Xai’s Grok 4.
Carlsen and Howell turned between serious comment and a roast when Grok’s performance came out as something erratic. In the first game it quickly sacrificed its bishop and then started shopping with pieces as if it were busy going home. Things did not improve in the next game for Grok.
“[Grok] is like the one guy in a club tournament who has learned theory and literally doesn’t know anything else, “Carlsen said during the other game.” Do the worst tabs after that. “
Grok’s performance was so off-the rails that Carlsen assessed it about 800 ELO or just over a beginner. He gave O3 a modest but respectable 1200, in the middle of most hobby players. Although O3 did not play brilliantly, it did not need. It played solid chess. It did not dazzle pieces. It converted its advantages and performed the classic chess movements.
“O3 is pretty ruthless in conversions; it looks like a chess player. Starting looks like it was learning a few opening movements and knowing the rules, but not much more.” Carlsen said. “Groks movements are chess -related movements. They just came at the wrong time and in strange sequences.”
Chess ai
The shake was not the most important point in the tournament despite its prominent character. It was about how AI models for general purposes handle events with strict rules such as chess games. It turns out they’re not big, but O3 is the best of the limited test. When AI gets embedded in everything, the ability to follow rules and spot patterns becomes important. Chess is a unique transparent way of observing it. You either made the right step or you didn’t. When a model plays well, you can see the logic; Otherwise, Queens falls as dominos and the game gets as confused as it metaphor.
Chess is a window into how well an AI can plan, evaluate options, avoid catastrophic errors and remain logically consistent. If you are throwing a queen away because it does not intervene in long -term consequences, what can it do in a legal document or when booking traveling?
The fact that the final was between Openai and Xai added some drama with Sam Altman and Elon Musk on Loggerheads in public. The Chess final did not solve the battle between them, but it gave Openai a PR -victory within the public view and a limited but very real compliment from Magnus Carlsen.
