The Kingdom of Saudi Arabia recently set ambitious targets in its national transformation program and Vision 2030 to move away from oil dependence and redirect oil and gas exploration to other higher-value uses to supply 10% of its energy demand from renewable sources. The incorporation of solar energy into the grid becomes important due to its ever-increasing demand growth. This paper proposes to use machine learning techniques to predict daily GHI in some cities in Saudi Arabia. The paper compares the prediction of GHI using Artificial Neural Networks (ANNs), Support Vector Machine (SVM), Random Forest, and Random Tree. The data used in this study uses attributes such as variable weather and solar irradiation data provided by KACARE. A comparison between actual and predicted GHI revealed that Random Forest performed better than the rest of the machine learning techniques with a root mean square error (RMSE) and a correlation coefficient of (45.560, 0.990) respectively compared to ANN (106.709, 0.984), SVM (203.79, 0.77), Random Tree (80.326, 0.968). The significance of the study relies on its ability to predict solar GHI for sustainable integration of PV systems the electrical grid and help operators manage the power generated more efficiently.