This chapter focuses on how deep learning-based generative algorithms are crucial for developing trustworthy AI datasets, especially in oncology. We start by outlining the basics of synthetic data generation in cancer research, introducing the concepts and techniques that enable the creation of realistic and diverse datasets. Following this, we highlight key successes and applications from the literature, showcasing the practical benefits and potential of these algorithms. We then evaluate the trustworthiness of synthetic data, discussing how to assess its quality and reliability for use in clinical and research settings. In the last section, we will identify challenges and opportunities that lie ahead, emphasizing the role of innovation in expanding the use and impact of synthetic data in cancer research.