Stealth coatings，by regulating radar waves，infrared radiation，visible light，and laser signals，were widely applied in military equipment and advanced technological fields. However，the design of stealth coatings involved time-consuming experiments due to thecomplexity of material selection and processing parameters. To address these limitations，data-driven coatings design methods had attracted increasing attention. This review highlightedrecent advances in stealth coatings design based on machine learning. It summarized the maintypes of stealth coatings，including radar-absorbing，electromagnetic shielding，infrared stealth，and composite stealth coatings，while discussing the challenges of traditional designmethods. The review introduced data-driven stealth coatings design approaches，demonstratinghow data preprocessing and feature extraction strategied optimize model inputs. It underscored the significance of high-quality databases，model interpretability，and multi-objective optimization. Additionally，research cases were presented where machine learning had been applied in performance prediction，material screening，structural design，and inversed optimization of stealth coatings. Finally，recent data-driven research advancements in functional coatings across various fields were explored，providing valuable insights into the intelligent design of future stealth coatings.