Development of Digital Forensics Technique Based in AI to Identify Deepfake Video and Image

The rapid advancement of artificial intelligence, particularly deep learning, has enabled the creation of highly realistic synthetic media known as deepfakes. These manipulated images and videos pose serious threats to digital security, privacy, and societal trust by enabling misinformation, identity fraud, and cybercrime. Traditional digital forensic techniques are increasingly challenged by the sophistication of deepfake generation methods, which can mimic facial expressions, voice patterns, and visual textures with high accuracy. Consequently, there is an urgent need for advanced forensic frameworks capable of detecting and analyzing deepfake content effectively. This research presents the development of an artificial intelligence-based digital forensic technique for identifying deepfake images and videos. The proposed framework leverages machine learning and deep learning models such as Convolutional Neural Networks, Recurrent Neural Networks, and hybrid architectures to analyze spatial and temporal inconsistencies in media content. The methodology integrates feature extraction techniques, including facial landmark analysis, frequency domain analysis, and texture-based detection, to improve detection accuracy. Recent studies have demonstrated that AI-based detection models can effectively identify subtle artifacts introduced during deepfake generation, such as inconsistencies in lighting, blinking patterns, and compression artifacts [1][4]. However, these models often face challenges related to generalization, dataset bias, and adversarial attacks. This research aims to address these challenges by proposing a robust and scalable forensic framework that combines multiple detection techniques and ensures interpretability. The expected outcome of this study is the development of a reliable and efficient deepfake detection system that can be integrated into digital forensic workflows and cybersecurity systems. The research contributes to enhancing digital trust and combating misinformation by providing advanced tools for identifying manipulated media.