Termite attacks result in substantial economic and environmental losses annually. Current prediction methods are both labor-intensive and costly. To address this issue, the study presents "anAI," a mobile application that employs deep learning and machine learning models to forecast termite infestations, identify wood types, and assess the likelihood of attacks using environmental data. The research utilized three deep learning models—MobileNetV2, AlexNet, and CNN—for image classification, with CNN achieving the highest accuracy rates. Additionally, machine learning models such as Artificial Neural Networks (ANN) and Support Vector Machines (SVM) were employed to predict termite attack likelihood, with SVM proving superior. Logistic regression classified likelihood into three levels: certain, likely, and unlikely. The study collected 181 wood samples, augmenting them to 506 images, and achieved satisfactory accuracies for termite infestation and wood type identification. "anAI" delivers real-time predictions, facilitating proactive termite management by utilizing environmental parameters like temperature, humidity, and wood moisture. This research underscores the importance of integrating advanced intelligent systems into pest management, enabling early detection and mitigation of termite infestations, thereby minimizing property damage, and supporting sustainable practices.