This study focuses on the development of an AI model to predict the next variants of the SARS-CoV-2 virus based on genomic data. Leveraging a dataset comprising a wide range of SARS-CoV-2 variants, including the Alpha, Beta, Gamma, Delta, Epsilon, and Omicron variants, we employ Artificial Intelligence (AI) algorithms to train a model capable of identifying patterns and mutations within the viral genome. Furthermore, we emphasize the significance of the Spike protein region, given its relevance to vaccine development. By treating the Spike protein sequences as 2-dimensional images, we apply image recognition techniques commonly used in AI research to analyze and extract meaningful insights from these protein sequences.We also highlight the status of AI applications in genomics, noting previous studies focused on binding affinity prediction and clustering analysis of Spike proteins. However, no existing AI models have addressed the prediction of future variant sequences. Consequently, our study aims to bridge this gap by developing an AI model with the potential to forecast the sequence of forthcoming SARS-CoV-2 variants. This research contributes to our understanding of viral evolution and assists in the proactive development of strategies to combat the evolving COVID-19 landscape.