Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{200623,
        author = {Devi Krishna P.V and Angelin Bibesha S.M and Diya Jithu A.H and Subhashini S},
        title = {DEEP FAKE DETECTION WITH EXPLAINABILITY},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {1636-1642},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=200623},
        abstract = {Deepfake technology has advanced significantly in recent years, enabling the creation of highly realistic synthetic images and videos that can mislead viewers and undermine digital trust. These manipulated media pose serious threats to information integrity, social stability, and security across domains such as journalism, law enforcement, and education. Although numerous deep learning-based detection systems have been developed, most operate as black-box models that provide only binary classification results without offering insights into their decision-making process. This lack of transparency reduces user confidence and limits real-world applicability in critical scenarios. To address this challenge, the proposed system presents an explainable deepfake detection framework that combines high-accuracy detection with interpretability. The system leverages advanced convolutional neural networks to identify manipulated media, while integrating explainability techniques such as Gradient-weighted Class Activation Mapping (Grad-CAM), feature attribution methods, and artifact-based analysis. These techniques enable the system to visually highlight suspicious regions in images or video frames and generate human-readable explanations that describe the basis of detection. By providing both visual and textual justifications, the system enhances transparency, interpretability, and trustworthiness. This approach not only improves user confidence but also serves as an educational tool for understanding deepfake characteristics. The proposed framework contributes to the development of reliable and explainable artificial intelligence solutions, promoting responsible usage and wider adoption of deepfake detection technologies in real-world applications.},
        keywords = {Deepfake Discovery, resolvable AI, CNN, Transformer, Grad- CAM, SHAP, Digital Forensics},
        month = {May},
        }