Fabricated Reality: Understanding the Landscape of Artificially Generated Disinformation in the Age of Deepfakes

This chapter maps the landscape of artificially generated visual disinformation, arguing that the rise of deepfakes and AI-generated imagery demands a fundamental rethinking of visual literacy. It traces the lineage of visual manipulation from Photoshop in the late 1980s to today’s accessible smartphone filters and generative tools (DALL-E, Midjourney, Stable Diffusion), contending—after Swerzenski and Messaris—that simplistic “real or fake” labeling is inadequate and that literacy must instead address the processes producing images. Drawing on Baudrillard’s simulacra, Flusser, Boorstin, and theorists of image interpretation (Barthes, Berger, Sontag), it frames AI-generated images as copies without originals that sever the long-assumed link between photography and reality, illustrated by cases such as Boris Eldagsen’s award-winning AI image and the fabricated photo of Pope Francis in a puffer coat. The chapter stresses that visual disinformation is understudied yet especially potent, since audiences process images less critically, and documents how even professional newsrooms republish decontextualized or doctored visuals under breaking-news pressure. It surveys verification practices and detection methods—reverse image search, metadata and light/shadow analysis—alongside common manipulation techniques (copy-paste, montage, retouching, misleading cropping) and their evasion. It concludes that visual disinformation differs fundamentally from textual forms, exerts stronger effects, and requires distinct analytical approaches, while current detection models remain limited.