Facial age progression is a critical task in forensic investigations, aiding in the identification of missing persons, reconstructing potential appearances in cold cases, and enhancing law enforcement tools. Traditional methods often suffer from identity loss, unrealistic texture generation, or require extensive manual intervention. This paper introduces an AI-driven approach utilizing StarGAN v2, a generative adversarial network (GAN)-based model, to achieve realistic, identity-preserving facial aging. The proposed framework employs a generator with an encoder-decoder structure, leveraging residual blocks and adaptive instance normalization to ensure age-conditioned transformations while maintaining facial identity. A style encoder extracts age-specific features from reference images, allowing the model to generate highly realistic aged representations with domain-specific accuracy. The adversarial learning process, enforced by a discriminator network, enhances image quality by distinguishing real aged faces from synthetically generated ones. To validate the effectiveness of our approach, we conduct extensive experiments on publicly available facial datasets, ensuring generalization across diverse ethnicities, age groups, and facial structures. Forensic case studies demonstrate the practical applicability of our model, with perceptual similarity metrics and human assessment studies indicating superior identity preservation and visual accuracy compared to traditional methods. This research underscores the potential of GANs in forensic applications, offering an automated, scalable, and adaptable solution for facial age progression. By addressing key challenges in aging synthesis, our approach enhances investigative methodologies, paving the way for advancements in forensic facial reconstruction.