Structural corrosion due to chloride ingress is considered to be the greatest threat to buildings. Structures such as bridges, roads, infrastructure, and harbors are exposed to chloride-rich environments, either through anti-freeze salting or from the natural environment. In the marine environment, tidal and wave areas are considered to be at high risk of corrosion. The corrosion process is highly dependent on the chloride concentration in the concrete structure. Therefore, the determination of chloride concentration in concrete is meaningful and reliable. In this study, a Gradient Boosting (GB) machine learning model is proposed to predict chloride content in concrete. A dataset of 325 experimental results was collected from international literature. The results of the model are programmed and run in the Python platform, the authors use the tool of the Gradient Boosting model to predict the output parameters of chloride content in concrete. The results show that the proposed model can predict the chloride content in concrete simply and quickly, helping design engineers to predict the chloride content in advance based on the input parameters