MAD'25: 4th ACM International Workshop on Multimedia AI against Disinformation

In recent years, it has become increasingly difficult to verify the accuracy of media. Advancements in AI generated content-such as text, images and videos created by foundation models-have made it progressively harder to distinguish authentic multimedia content from synthetic data. AI-powered tools and methods are commonly used to create and augment disinformation, including to spread it widely in automated ways, or to impersonate and defame individuals, resulting in undermined trust in the online information, and ultimately posing a significant threat to society and democracy. This creates a growing need for tools that can combat disinformation and verify the authenticity of multimedia content. The MAD'25 workshop aims to bring together researchers and practitioners from different backgrounds, under the shared goal of combating disinformation through the use of AI. In its fourth edition, the workshop aims to foster an environment for exploring innovative ideas and sharing experiences. The research areas of interest include the detection of AI manipulated or generated content, as well as the study of how disinformation spreads and its broader societal impact. Given the crucial role of multimedia, the workshop places particular emphasis on the integrated analysis of multiple modalities-such as text, audio, and video-as combining different forms of content could enhance verification efforts.