<p>Clustering plays a pivotal role in unsupervised learning, enabling the identification of meaningful patterns and structures in datasets without prior knowledge. It is widely applied in various real-world applications, including but not limited to customer segmentation, image processing, and bioinformatics. Traditional clustering methods struggle with high-dimensional and nonlinear data distributions, which has motivated researchers toward the Nature-Inspired Optimization Algorithms (NIOAs) as a better option for complex problems. This study presents the African Vultures Optimization Algorithm (AVOA) for clustering analysis, marking the first comprehensive examination in the clustering domain. This work demonstrates the efficiency of AVOA with substantial experimental data evaluated with twelve benchmark UCI and synthetic datasets using eight established NIOAs. Extensive experiments show that AVOA consistently achieves lower intracluster distances and superior convergence behavior across most datasets. Furthermore, the performance of the proposed approach has been quantitatively evaluated using standard clustering validity metrics, namely the Davies-Bouldin Index (DBI), Calinski-Harabasz Index (CH), Silhouette Index (SI), and CS Index, along with execution time (ET) to assess computational efficiency. These metrics are used consistently to compare the proposed method with the other algorithms considered in this study. In addition, we have also included a high-dimensional and highly imbalanced dataset, SECOM, for performance evaluation. Experimental findings and statistical analyses using Friedman, Iman-Davenport, and Holm tests show the significant superiority of AVOA, and convergence graphs and box plots highlight the robustness, scalability, and stability of the proposed strategy in clustering.</p>