Ask AI what goes with chicken and the answer depends on whether it learned from recipes or molecules

London-based startup Kaikaku.AI released "Epicure," three AI models trained on 4.14 million recipes and a flavor database, that separately determine whether an ingredient fits a recipe or is chemically related. The chemistry-based model classifies taste and nutritional value better than recipe-based alternatives, despite never having direct access to that information. The distinction matters because it reveals how AI's training data fundamentally shapes its recommendations for everyday tasks like food pairing.