The paper introduces AMDLmodel, a deep learning technique based on a convolutional neural network (CNN) for detecting and classifying Android malware. The model uses the Drebin dataset, which contains 215 feature attributes from 15,036 applications, including 5,560 malware and 9,476 benign applications. The CNN architecture includes convolutional layers, pooling layers, and fully connected layers, with ReLU activation functions. The model is trained and tested using various parameters such as filter sizes, number of epochs, learning rates, and layers. The evaluation metrics include accuracy, precision, recall, and F1-score, with the AMDLmodel achieving an accuracy of 99.92%. The study compares the AMDLmodel with other state-of-the-art techniques, demonstrating its superior performance in malware detection. The paper also discusses the limitations and future work, emphasizing the need for broader datasets, enhanced model interpretability, and real-time detection.The paper introduces AMDLmodel, a deep learning technique based on a convolutional neural network (CNN) for detecting and classifying Android malware. The model uses the Drebin dataset, which contains 215 feature attributes from 15,036 applications, including 5,560 malware and 9,476 benign applications. The CNN architecture includes convolutional layers, pooling layers, and fully connected layers, with ReLU activation functions. The model is trained and tested using various parameters such as filter sizes, number of epochs, learning rates, and layers. The evaluation metrics include accuracy, precision, recall, and F1-score, with the AMDLmodel achieving an accuracy of 99.92%. The study compares the AMDLmodel with other state-of-the-art techniques, demonstrating its superior performance in malware detection. The paper also discusses the limitations and future work, emphasizing the need for broader datasets, enhanced model interpretability, and real-time detection.