The gravitational search algorithm offers a robust framework for clustering tasks by leveraging principles inspired by Newtonian gravitation, where data points act as masses influencing each other through simulated gravitational forces. This approach enhances clustering accuracy by capturing multidimensional relationships, particularly in complex datasets where conventional methods like K-means struggle with local optima. Improvements through age diversity address GSA’s tendency toward premature convergence, a limitation observed in feature selection and optimization tasks. By integrating age-based population management, older solutions can be systematically replaced or mutated to maintain genetic diversity, preventing stagnation in local minima. This strategy complements existing adaptive enhancements such as dynamic gravitational constants and distance recalibrations, which improve inter-particle information exchange Empirical validations using synthetic datasets, CEC2017 benchmarks, and gene expression data confirm that age-diversified GSA variants achieve higher silhouette score (+18–22%) and faster convergence than classical implementations. ADGSA consistently outperforms all other clustering algorithms across the evaluated datasets, with notable percentage improvements. ADGSA still manages steady leads with percentage improvements ranging from (5.56%) to (41.03%) against GSA and (2.70%) to (27.91%) against SAGSA, showcasing its robustness even among more advanced techniques.