Photorealism versus photography. AI-generated depiction in the age of visual disinformation

Abstract

ABSTRACT In the spring of 2023, we witnessed a breakthrough in the development of AI-generated images accessible to the general public. Pictures of Pope Francis wearing a stylishly long, white puffer jacket or driving a motorcycle down a busy street went viral. The same thing happened with AI-generated pictures fabulating over what the press coverage of an imminent arrest of former US President Donald Trump would look like. Amnesty International used AI to generate images to mark the second anniversary of police violence against protesters in Colombia. Boris Eldagsen turned down an award for Best Creative Photograph from the Sony World Photography Awards, announcing that it had been generated with AI. Critical reactions in the public were not long in coming. The use of AI to generate images was discussed with sometimes shrill words and phrases such as fake, fraud, a threat to photography’s credibility, and fake news. This article seeks to intervene in this crisis-oriented debate by proposing three analytical moves: First, we need a concept of photorealism that is kept separate from the idea of photography. Secondly, we need a conceptual distinction between two basic functions of photography: depiction and detection. In addition to this primary distinction between image functions, the article proposes a third move to introduce a function-oriented genre concept. Through an interdisciplinary approach to photorealism, photography, and genre, these three analytical measures are presented and examined step by step and discussed through analyzes of concrete and recent examples of AI-generated and AI-enhanced publicly available images in today’s society. Today’s crisis-oriented public debate about AI images serves neither democracy, art, journalism, nor photography. The purpose of the article is to contribute a simple and, at the same time, useful analytical tool in these discussions about the relationship between photography and current and future image technologies.