Deep learning algorithms have become so powerful as computing power has increased that it has become much easier to create indistinguishable human-synthesized videos, dubbed DeepFake. It’s easy to imagine scenarios in which these realistic Deepfakes are used to commit political pranks, staged terrorist assaults, revenge pornography, and blackmail. This challenge describes a revolutionary deep learning-based technique for detecting genuine AI-generated videos from forgeries. Our method can automatically recognise surrogate and recurring deepfakes. We’re using artificial intelligence to try to counteract it. Our solution use a Deep Convolutional Neural Network to extract framelevel data and then train those features. A recurrent neural network with long-short-term memory used to determine if a picture has been edited. A huge set of balanced blended datasets was created by combining the numerous datasets available in order to recreate real- time events and enhance model performance using real-time data. A simple and dependable way for illustrating how a system might produce competitive performance. An interactive Graphical User Interface (GUI) is also developed for easy access.