The most complex disease in the world, oral cancer is caused by tumour that grow in the tonsils, salivary glands, neck, face, and mouth. Its late identification is causing an increase in the death rate worldwide. The early detection of oral lesions can save more lives by providing prompt and early treatment by doctors which needs advanced investigation methods and develop predictive algorithms for cancer management. This survey paper discusses the existing research work carried out using machine learning and deep learning mechanism and evaluate its limitations towards classification accuracy, predictive analysis performance. This article discusses survey analysis, research gap identified. Oral cancer remains a significant global health concern, necessitating advanced diagnostic and prognostic tools for effective management. It is one of the complex diseases in the world. The rate of the death is increasing all over the world due to this type of malignancy. Early diagnosis of malignancy is essential to avoid risk of disease. Earlier detection of oral lesions aids surgeons to provide proper treatment to save lives. This survey paper comprehensively reviews recent advancements in using deep learning and machine learning techniques for the detection, classification, and prognosis of oral cancer. The review synthesizes findings from diverse studies, including CNN-based frameworks for dysplastic tissue classification, novel deep learning approaches for oral cancer diagnosis, and automated systems integrating AI for early detection. Key studies highlight the effectiveness of techniques such as Deep Belief Networks, convolutional neural networks (CNNs) like DenseNet169, and innovative methodologies like digital processing of CT images and non-invasive photographic analysis. These approaches consistently demonstrate high accuracy rates comparable to expert assessments, paving the way for enhanced clinical decision-making and patient outcomes. The paper also discusses challenges, such as dataset heterogeneity and model generalization, and proposes future research directions, including data fusion algorithms and integration of advanced imaging modalities. This survey underscores the trans-formative potential of AI-driven technologies in improving oral cancer detection and management globally. Search for recent advancements in deep learning techniques used for oral cancer detection. Find information on the effectiveness of convolutional neural networks (CNNs) like DenseNet169 in classifying dysplastic tissue related to oral cancer. Research the application of Deep Belief Networks in the diagnosis of oral cancer. Explore methodologies involving digital processing of CT images for early detection of oral cancer. Investigate the use of non-invasive photographic analysis combined with machine learning for oral cancer diagnosis. Identify the challenges associated with using deep learning and machine learning for oral cancer, such as dataset heterogeneity and model generalization. Find information on proposed future research directions in this field, including data fusion algorithms and the integration of advanced imaging modalities.