<p>Predictive maintenance (PdM) is a critical tool in aviation, promoting sustainability, safety, and cost-effectiveness. However, a significant challenge in implementing predictive maintenance framework is limited run-to-failure data due to frequent preventive maintenance. Generative AI (GAI) offers a promising solution to this challenge by generating synthetic data, enabling more accurate predictions of aircraft system health. The recent surge in scientific publications exploring GAI’s potential for aviation predictive maintenance emphasizes the need for a comprehensive review. This research addresses this gap by conducting a Systematic Literature Review (SLR), employing an active learning open-source tool to thoroughly analyze papers sourced from four scientific databases. These papers focus on predictive maintenance of aircraft systems through generative AI models. The findings of this review examine various aspects of GAI-powered predictive maintenance, including its objectives, the diversity of models employed, areas of application, datasets used for model validation, and the prevailing challenges and emerging trends. The review identified that most commonly used GAI models in aircraft PdM are generative adversarial networks, variational autoencoders or combination of these models. However, knowledge sharing, model integration and specificity are the key challenges associated with the implementation of these models. This strong knowledge-base will be a valuable resource for researchers, engineers, and practitioners aiming to improve their knowledge and implementation of GAI-driven models for predictive maintenance in the aviation industry.</p>