Accurate determination of water saturation is fundamental for predicting the amount of hydrocarbon fluid in reservoir, apart from monitoring fluid saturation changes whithin the reservoir using 4D seismic technology in the most efficient manner. In fact, although several techniques can be employed to predict water saturation, the well log based approaches have emerged as the most widely applied methods. Nevertheless, even though these techniques offer rather exceptional vertical resolution, it is a fact that lateral estimation is still an issue. As such, this study proposes a methodology that estimates the water saturation based on pre-stack seismic information by using artificial intelligence (AI) technologies, such as artificial neural network (ANN), adaptive neuro-fuzzy inference systems optimized by genetic algorithm (ANFIS-GA), as well as support vector machine (SVM). Therefore, the seismic data would go through a process known as seismic inversion and then transformed into seismic attributes, represented as SQp and SQs, this transformation is adherent to attenuation concept, which is associated to the rock physics. On top of that, data from two wells in Malaysian hydrocarbon field were employed to build the estimation models of ANN, ANFIS-GA, and SVM and then third well data was used as blind well to test unseen data. Furthermore, both SQp and SQs had been employed in the models as input variables, whereas the water saturation functioned as output. Other than that, several performance indices of statistical had been incorporated in this study so as to perform a comparison between the performances exerted by the selected models. As a result, the findings indicated thatalthough all models showed an acceptable agrrement with the experimental data to some extent, the ANFIS-GA based model had a better performance over other models. The present study show that the AI technology can be incorporated in water saturation prediction workflow from pre-statck seismic data which result in superior accuracy and performance in fluid flow calculations and time lapse seismic analysis. © 2018, Offshore Technology Conference.