This thesis explored the use of ensemble learning techniques for the detection of manipulated digital faces. The research aimed to develop an approach to generalize digital face manipulation detection by using ensemble learning methods to combine the predictions of multiple classifiers in a hierarchy model. The study used various datasets of malicious and benign manipulated, and bonafide digital face images. Many of the manipulated face images and all of the bonafide images were taken from existing databases, with proper references, while the rest were generated using OpenCV and custom functionality. The results showed that the ensemble learning approach achieved good recall, precision, as well as APCER and BPCER. The research has significant implications for the detection of digital face manipulation in various applications, including forensics and security. The study demonstrates the effectiveness of ensemble learning techniques for detecting digital face manipulation and highlights the importance of using a generalized approach to achieve higher accuracy and reliability in the detection process.The results show that while using an ensemble model for PAD yields prominent results, there is still a long way to go before the model can be used in commercial options.