The rapid-fire increase in videotape content across digital platforms has boosted the need for effective results to decry manipulated videos. These include deepfakes and other phonies, which pose significant pitfalls to digital trust and security. This paper introduces a new approach, the" AI videotape Authenticity Checker," which employs a mongrel deep literacy frame. The model utilizes Convolutional Neural Networks( CNNs) for spatial analysis and Long Short-Term Memory( LSTM) networks for temporal analysis, furnishing a robust result for detecting fake vids. By preprocessing videotape data to regularize quality and format, the system ensures high trustability and scalability. Experimental results on standard datasets demonstrate a delicacy of 92, a perfection of 90, a recall of 91, and an F1-score of 90.5, showcasing its eventuality for real-world operations. This scalable and effective tool represents a critical advancement in videotape phoney discovery.