This market will resolve yes if an LLM/Agent system is able to beat a 2000 rated elo chess player on chess.com or lichess equivalent rating of 2000.Rules:The AI is allowed to use images,reason,search online for moves, but not use chess engines to make moves.The AI is allowed to continue if it makes mistakes. For resolution sharing the chat the game is played in is necessary. Gamemodes can be rapid(10 to 60minutes)
Update 2025-07-20 (PST) (AI summary of creator comment): The creator has clarified the rules:
The AI is not allowed to ask a human for help.
Update 2025-07-20 (PST) (AI summary of creator comment): - The human player must play at their best and is not allowed to intentionally lose the game.
Update 2026-02-05 (PST) (AI summary of creator comment): If a suitable 2000 Elo volunteer player cannot be found to play against the AI model, the market will resolve based on the AI playing against a bot with an approximate Elo translation to a 2000-rated player.
1,000People are also trading
@traders This market will close in almost a year so and right now no frontier model has come close to this yet. Also we have to find 2000 elo volunteers to play against this supposed end of 2026 or early 2027 ai model. If we can't find anyone then we will have to play against a bot and do a roughly elo translation to a player.
@traders has anyone played against ChatGPT agent? The agent searches for moves, and you place the pieces that the ChatGPT agent plays. I am currently testing, and so far, there have been no wrong moves. The only drawback is the usage limit of 40 messages per month, which limits the agent's capabilities to approximately 40 chess moves.
@rada maybe I should I clarify the nuances of the question you must play at your best and plus you must setup the time constraints if you are going to play against the LLM yourself in chess.com that means you must move the chess pieces to the position it gave you or if it is agent system it could just move it by itself