The field of digital image integrity and security is rapidly evolving, with a focus on developing innovative solutions to protect against misinformation, fraud, and intellectual property theft. Researchers are exploring new techniques for detecting and localizing image forgeries, as well as developing robust watermarking methods to ensure the authenticity and ownership of digital images. The use of large vision-language models and multi-agent frameworks is becoming increasingly popular in this area, enabling more accurate and efficient detection of AI-generated images and image manipulations. Furthermore, the development of novel metrics and datasets is facilitating the evaluation and improvement of digital image integrity and security systems. Notable papers in this area include: UniShield, which proposes a novel multi-agent framework for unified forgery image detection and localization, achieving state-of-the-art results and superior practicality. SpecGuard, which introduces a spectral projection-based approach for robust and invisible image watermarking, outperforming existing state-of-the-art models in terms of invisibility, capacity, and robustness.