The vulnerability of wireless communication to attacks, combined with the highly dynamic nature of VANETs and the stringent requirements for accuracy and speed, necessitates the development of specialized solutions for these networks. Trust systems have emerged as an important area of interest in both academic and industrial contexts, especially considering the limitations of cryptographic methods that primarily address external threats. While trust systems utilizing machine learning have been used for some time, the application of deep learning in this field is relatively new. This study presents an analysis of the algorithms used and their application areas, focusing on recent developments in the integration of deep learning approaches within trust systems. It is anticipated that the findings of this study will contribute to the future development of deep learning-based trust systems.