OpenAI’s o3 Dominates Elon Musk’s Grok 4 to Win Kaggle AI Chess Tournament

OpenAI's AI model, o3, triumphed over Elon Musk's AI Grok 4 with a perfect 4-0 in the final match of the Kaggle AI Chess Exhibition Tournament. The tournament featured a series of chess matches with various top AI models competing against one another.

Grok 4 looked strong before the final, it had beaten all other competitors easily.

However, Grok made a lot of blunders in the final, while O3 played consistently and capitalized to win each game.

In the third-place match, Google's Gemini 2.5 Pro defeated OpenAI's o4-mini by a score of 3.5 to 0.5.

This event was special because it was the first major chess competition using general-purpose AI language models like ChatGPT. According to experts, these AIs have not yet reached human grandmaster levels. Magnus Carlsen, the world chess champion said he rated Grok's playing strength at about 800 and his o3 at about 1200 while Carlsen's best rating was 2882. 

Elon Musk said Grok's chess rating was just a "side effect" since his would not focus much on chess. 

The tournament demonstrated how AI is developing reasoning and playing games that require complexity such as chess. It also showcased the rivalry between OpenAI and Musk's xAI company, who used to work as partners but now appear to be competing. 

All of the games and commentary are posted on-line for those interested in watching.