Aesthetics and AI Image practices

Join us on Wednesday 30th March 2022 at 15.00 at Borough Road Gallery, LSBU, or online at https://centreforthestudyof.net/conversations, for our next research event with Aurora Hoel, Gabriel Pereira, and Lotte Philipsen. All welcome.

The presentation and following discussion focus on the ability (or not?) of aesthetic theory to analyze AI image practices. Traditionally ‘aisthesis’ implies subjective sensuous encounters with an object/phenomenon, whereas ‘poiesis’ implies creation of objects/phenomena, but in AI image practices (like synthetically generated images and automated object detection) traditional roles of viewing subject, viewed image-object, creating subject, and created image-object seem shuffled, mixed and partly dissolved. In this case, what kind of aesthetic theory is useful for analyzing concrete AI image practices and how would such analysis differ from cultural or technical analysis? In their joint presentation, Aurora Hoel, Gabriel Pereira, and Lotte Philipsen offer some tentative suggestions.  

Aurora Hoel is professor of media studies and visual culture at the Norwegian University of Science and Technology (NTNU). Her research focuses on operational images and media, with a special emphasis on technical images and machine vision.

Gabriel Pereira is a Visiting Fellow at the London School of Economics and Political Science (UK), funded by the Independent Research Fund Denmark International Postdoc grant. His research focuses on critical studies of data, algorithms, and digital infrastructures, particularly those of computer vision.

Lotte Philipsen is associate professor in Art History with Aarhus University. Her research focuses on image theory, contemporary art, and digital aesthetics. She is PI on the project New Visions: Image cultures in the era of AI (2023-2026) and leader of an upcoming Center for Aesthetics of AI Images (AIIM) at Aarhus University.

The attached image is automatically generated via text-input on the portal DeepAI.org: https://deepai.org/machine-learning-model/text2img. The text input was “aesthetics of ai images”.