A LIGHTWEIGHT AI MODEL FOR DEEPFAKE DETECTION IN CYBERSECURITY

The deepfake technology has become a major security threat in cybersecurity because it can produce very realistic fake images and video that can be misused and cause identity theft, fraud, and misinformation. The following paper describes a simple model of detecting deepfake images with the help of a simple convolutional neural network (CNN) based on artificial intelligence. The suggested model is expected to be performance and computationally efficient to be applicable in real-time and resource-constrained settings. The model is trained using a balanced set of real and counterfeit images using simple preprocessing methods of resizing, normalization, and data augmentation. The experimental findings indicate that the model has an accuracy of about 91, with even balanced value of precision and recall. The results indicate that a lightweight model can be used to detect deepfake materials with fair precision. Nevertheless, the research also attributes drawbacks like low work on high-quality deepfakes and reliance on the size of the dataset. In general, the given approach offers a viable solution to the improvement of the cybersecurity systems but may be enhanced to more sophisticated situations of detection. Keywords: Deepfox Detection, Artificial Intelligence, Cybersecurity, Convolutional Neural Network, Image Classification, Lightweight Model, Fraud Detection, Digital Security