Bridge after a period of operation need to have assessments of the actual operating loads of these works. The current practices in Vietnam are usually applied according to the US bridge assessment guidelines (The Manual for Bridge Evaluation (MBE). With the traditional assessment, it is necessary to determine the bearing capacity of the structure and the effects caused by the dead and live loads. The above calculation work requires in-depth knowledge and clear understanding of its components. Today, with the development of science and technology, not only computers and smartphones are becoming more and more popular, the application of artificial intelligence has entered most areas of life. Therefore, the application of artificial intelligence in bridge load rating is no longer a strange job. The article builds an artificial neural network and applies it to the load rating of steel girder bridges, the results obtained by the proposed method are compared with the traditional method, the result of applying another artificial neural networks and give results with high agreement.