Estimating dew point pressure is very important for gas reservoirs evaluation and simulation. The dew point pressure is usually measured from collected fluid samples. However, sometimes, these measurements are not available. People tried to estimate the dew point pressure using explicit methods like empirical correlations or using iterative methods like equation of state. Empirical correlations are fast and easy to use but usually they are not very accurate. EOS, although more accurate, it is more expensive computationally and needs to be calibrated to existing experimental data. Artificial Neural Networks has been more popular recently for complex input-output mapping. It has been used in the literature to predict bubble point pressure in oil fields and good results has been reported. In this paper, we designed Artificial Intelligence models to estimate the dew point pressure in Saudi Arabia gas condensate fields. We used data from 98 PVT reports to train, validate, and test the models. The results are discussed in this paper.