Self-consolidating concrete (SCC) has revolutionized the concrete industry with its ability to reduce voids, flow easily into complex forms, and eliminate the need for mechanical vibration. However, one of the significant challenges with SCC is the corrosion of steel reinforcement caused by chloride penetration. Traditional methods for evaluating chloride infiltration typically involve 28-day tests, which often come too late for effective intervention. To improve prediction accuracy, the industry has explored various machine learning techniques for forecasting chloride permeability and compressive strength in advanced concrete mixes like high-performance concrete (HPC) and SCC. Among these, deep learning approaches, particularly artificial neural networks (ANN), have demonstrated superior accuracy. The proposed research work introduces ChloroNet-9 which is a framework designed to enhance the prediction of concrete strength and chloride permeability in sustainable green concrete. It comprises three key modules: Data Collection, Data Augmentation and Standardization, and Data Analytics and Prediction. Extensive data was collected from concrete mixes, focusing on properties like cement content, aggregates, and admixtures, which were then preprocessed using techniques such as Min–Max Scaling and SMOTE. ChloroNet-9 combines an artificial neural network (ANN) with random forest to prioritize important features, leading to highly accurate predictions. The ChloroNet-9 model significantly outperforms traditional methods, including XGBoost and regression algorithms, achieving an impressive 99% accuracy with very less mean absolute error of 3.06. By addressing the limitations of traditional testing methods, ChloroNet-9 emerges as a powerful tool for advancing the concrete industry, particularly in the context of sustainable construction.