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{201045,
        author = {Mrs. Malarvizhi and Chandru S and Aravindan N and Balachandru K and Anto Ashik D},
        title = {An AI-Powered Deepfake Detection System for Video and Audio},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {no},
        pages = {237-245},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=201045},
        abstract = {The rapid advancement of generative artificial intelligence has led to the proliferation of hyper-realistic synthetic media, commonly known as deepfakes. These manipulations pose significant threats to information integrity, cybersecurity, and social trust. This paper presents a novel AI-powered detection system capable of identifying forged content in both video and audio modalities. The proposed system integrates a dual-branch convolutional neural network with temporal attention mechanisms for video analysis and a spectrogram-based residual network for audio forgery detection. A multimodal fusion layer combines features from both streams to improve overall detection accuracy. The system ensures data integrity through cryptographic hashing and provides tamper-proof audit logging using blockchain technology. Experimental results on the FaceForensics++ and ASVspoof2019 datasets demonstrate a detection accuracy of 96.8% for video, 94.2% for audio, and 97.4% when using joint inference. The system also shows robustness against compression and noise artifacts. This work provides a practical, real-time solution for deepfake mitigation in digital media forensics.},
        keywords = {Deepfake Detection, Convolutional Neural Networks, Multimodal Fusion, Audio Forensics, Video Forensics, Generative AI Security, Blockchain, Temporal Attention.},
        month = {May},
        }