GENERIC WAR IMAGINARIES: AI-GENERATED IMAGES OF THE ISRAEL-GAZA CONFLICT IN THE ADOBE STOCK CONTROVERSY

Abstract

This paper investigates the changing documentality of AI-generated images and their role in the industrial production of war imaginaries. In doing so, we center our analysis on the paradigmatic case of Adobe Stock selling photorealistic AI images depicting the Israel-Gaza conflict. After the outbreak of the war in Gaza on October 7, 2023, the stock photo service Adobe Stock started to collect of AI-generated images produced by users depicting the conflict. Paying customers were then able to download and publish these images in both online and printed media. A public debate ensued when news media began using pictures from this collection, sparking discussion due to the lack of proper contextualization regarding their AI origin. The primary research question guiding this project is therefore: How does the popularisation and industrialisation of photorealistic AI-generated pictures alter the criteria by which we ascribe documentality to images employed in news reporting? To this end we conducted a mixed method research. In the first phase we collected and analyzed 55 articles related to the Adobe Stock case. The initial content analysis focused on framing of the news, in order to point out key themes and discourses surrounding the controversy. The second phase involves photo-elicitation interviews with photojournalists and photo editors, exploring the documentary and informational value of AI-generated images. Preliminary findings highlight the significance of metadata, the role of illustrative captions, and the aesthetics shaping representations of war and violence.