Rice cultivation is widespread globally, particularly in Asian countries, serving as a staple food for a significant portion of the world’s population. However, agricultural challenges such as rice diseases have plagued farmers for generations. Identifying these illnesses is essential, as severe infections may result in crop failures. Automated disease monitoring through rice plant leaf photos is essential for shifting from traditional, experience-based decision-making to data-driven approaches in agriculture. In this context, Artificial Intelligence (AI) has emerged as a promising instrument across multiple scientific fields, including plant pathology. This research presents a hybrid deep learning system developed for the automated identification of rice plant diseases with a curated collection of leaf images. The design utilizes MobileNetV2, a validated model, to extract profound information from input photos. These attributes are subsequently employed as inputs for various machine learning classifiers utilizing distinct kernel functions, implementing a rigorous 10-fold validation approach. The findings illustrate the efficacy of the suggested hybrid system, with an exceptional classification accuracy of 98.6%, a specificity of 98.85%, and a sensitivity of 97.25% utilizing a medium-sized neural network. The system not only outperforms traditional methods but also showcases computational efficiency and speed. Furthermore, the system is primed for testing with larger and more diverse datasets, paving the way for enhanced disease diagnosis in rice plants.