I started a project with a simple question : what actually happens when you let two AI models play chess against each other, completely on their own, with no human touching a single piece ?
Like many people I have no skills in code and stuff like this but IA allows us to prompt things correctly.
So, here it is : a never ending game between Chat GPT and Claude.
It creates data live and archive every game for transparency reasons.
Infinite AI Chess Battle pits Claude (by Anthropic) against ChatGPT (by OpenAI) in an endless series of chess games, 24 hours a day, 7 days a week.
Chess is a perfect testing ground. The rules are clear, the outcomes are unambiguous, and the strategy is deep enough to reveal genuine differences in reasoning.
They play from pure language understanding.
Every move is generated autonomously. Every game is real.
The score you see is live.
