Plant diseases pose a significant threat to global food security, particularly affecting tomato crops that are vulnerable to various pathogens. Despite advancements in disease identification methods, farmers continue to experience substantial decreases in yield due to delayed and imprecise diagnoses, along with inadequate treatment recommendations. This research aims to address this critical issue by developing an innovative Artificial Intelligence (AI)-based system that can detect tomato plant diseases and provide effective treatment suggestions. To achieve this, a Convolutional Neural Network (CNN) based on the InceptionV3 architecture has been trained using a comprehensive dataset of 11,000 tomato leaf images representing nine different diseases and healthy samples. The approach combines deep learning techniques for accurate image classification with natural language processing, leveraging OpenAI's GPT-3.5 Turbo model to generate customized treatment recommendations. The results demonstrate the exceptional performance of the model, with a training accuracy of 99.85% and a validation accuracy of 88.75%. Rigorous evaluation using confusion matrices and assessment metrics further confirms the model's high precision and recall rates for different disease categories, showcasing its robust generalization capabilities. Furthermore, the inclusion of an intuitive Streamlit interface enhances user experience and ensures practical applicability in real-world scenarios. This study makes a significant contribution to agricultural technology by providing a comprehensive solution that integrates precise disease detection with actionable treatment guidance. The developed system holds immense potential to revolutionize tomato crop management practices, potentially minimizing financial losses and promoting sustainable agriculture through targeted disease management strategies.