SYNTHETIC MEDIA DETECTION

Authors:

Deepika Yeredla, Dharmendra Kumar Roy, Vasanthi Rangu, Shirisha Ragate, Sandila Archana, Raye Sai Varun

Abstract—Synthetic media is digital content produced or modified by artificial intelligence (AI) techniques such as deepfake videos. The rapid progress of deep learning models has enabled the creation of highly realistic synthetic media. While these technologies are accelerating innovation in entertainment, education, and digital art, they pose significant ethical and societal risks, including misinformation, identity theft, reputational damage, and privacy violations. In this paper, an overview of synthetic media synthesis is presented, and various AI-driven detection methods are surveyed, with a specific focus on convolutional neural network (CNN)-based approaches. The proposed system aims to strengthen digital content verification and enhance media integrity by providing a reliable, automated framework that accurately distinguishes synthetic images from authentic ones.

Keywords – Synthetic media, Deepfake detection, Convolutional Neural Networks (CNN), Deep learning, Feature extraction, Image Classification.