Among 3D printing technologies, Fused Deposition Modelling (FDM) has occupied an important place due to its low cost and user-friendly nature, and it has been widely preferred. However, the moisture content of thermoplastic filaments used in this technology has had a direct impact on print quality. Moist filaments have led to issues such as weak interlayer bonding, surface defects, and reduced mechanical strength. Therefore, filament drying has been regarded as a critical step in the FDM process. Although conventional drying methods such as oven, hot air, vacuum have provided a certain level of solution, the manual determination of process parameters has caused efficiency problems in terms of energy consumption and processing time. In this study, the potential use of artificial intelligence (AI) technologies in filament drying processes for FDM has been examined in detail. Through machine learning, deep learning, and sensor integration, the real-time monitoring of filament dryness and the dynamic management of the drying process have been investigated. Moreover, various studies have demonstrated the positive effects of these systems on FDM print quality. The purpose of this study has been to present a comprehensive evaluation of AI-based drying approaches based on the existing literature and to establish a conceptual framework that will guide future research.