Abstract: CO₂ corrosion is very common in oil and gas pipelines, and it is one of the most serious corrosion problems. It is necessary to establish models or use methods to predict CO₂ corrosion, so as to understand and predict pipeline corrosion. Based on this, three common corrosion rate prediction models such as semi-empirical model, empirical model and mechanism model are introduced. Neural network, gray prediction, support vector machine and other CO₂ corrosion prediction methods are also introduced at the same time. According to the existing models and methods,the existing problems and future development directions are put forward.