The precise prediction of the dynamic modulus (E*) of asphalt concrete (AC) is essential for effective pavement design and performance assessment. The current study proposes a machine learning (ML) approach, utilizing the Random Forest (RF) algorithm optimized by the Fox optimizer, to predict E*. The model was developed and validated using a comprehensive dataset of 120 laboratory-tested AC mixtures, incorporating various input variables such as aggregate gradation, binder properties, and testing conditions. The resulting model, RF_FOX_10, exhibited high accuracy, achieving an R2 value of 0.989. The study also employed SHAP values to interpret the model’s predictions, revealing the relative importance of different input variables. The findings emphasize the significant influence of temperature and frequency on E*, while also demonstrating the contribution of aggregate gradation and binder properties. Developing a user-friendly graphical user interface further enhances the model’s practicality. This research significantly advances the development of efficient and reliable E* prediction methods, which directly contribute to enhancing pavement design and performance evaluation.