This market assesses whether a general-purpose AI image-generation model can consistently produce a chess board with pieces in their correct starting positions by July 31, 2025.

Resolution criteria:

• Success is defined as generating a proper chess board with all pieces in their correct starting positions 9 out of 10 times using a single prompt

• Prompt engineering is allowed, but must be a single prompt fed directly into the model with the first output being considered one attempt

• No other layers or post-processing allowed

• The model must be a general-purpose AI image generation or multimodal model (not specifically designed for chess)

References:

• https://manifold.markets/Shump/will-ai-be-able-to-generate-correct

• https://www.chess.com/blog/Rocky64/more-ai-chess-images

• https://deepdreamgenerator.com/tags/chessboard

• Update 2025-06-23 (PST) (AI summary of creator comment): The creator has clarified that models like Claude 3.7 will not be considered valid for this market, as they do not qualify as an image generation model in this context.

• Update 2025-06-25 (PST) (AI summary of creator comment): The creator is considering whether a prompt that includes an image is a valid method for satisfying the market criteria. This would be for use with a multimodal model that accepts both text and image inputs.

• Update 2025-06-29 (PST) (AI summary of creator comment): The creator has clarified that for the market to resolve to Yes, the prompt used must be a text prompt only. Prompts that include an image are not considered valid, even if used with a multimodal model.