The field of AI-generated media is moving towards greater emphasis on fairness and realism. Researchers are exploring ways to improve the accuracy and diversity of synthetic data, while also addressing concerns around bias and ethics. One of the key directions is the development of more sophisticated methods for generating synthetic faces, including the use of generative models that can create images with continuous and easy control of face identities and attributes. Another important area of research is the detection of deepfakes, with a focus on developing more generalizable and robust methods that can detect a wide range of unseen deepfake types. Noteworthy papers in this area include 'Rethinking Individual Fairness in Deepfake Detection', which proposes a framework for enhancing individual fairness in deepfake detection, and 'Seeing Through Deepfakes: A Human-Inspired Framework for Multi-Face Detection', which introduces a novel approach for detecting multi-face deepfake videos. Additionally, 'Vec2Face+' presents a generative model that creates images directly from image features, allowing for continuous and easy control of face identities and attributes, and 'Celeb-DF++' introduces a large-scale and challenging video deepfake benchmark for generalizable forensics.