AI-generated Images: Investigating the Aesthetic, Educational, and the Socio-technical Impact of Synthography

The proposed research aims to explore the epistemological, aesthetic, socio-semiotic, technical and educational implications of "synthographies," a term referring to images generated through Text-toImage (TTI) technologies such as DALL·E and Stable Diffusion. Conducted by an interdisciplinary team, this project will delve into the theoretical premises of AI-generated images and the consequences of the transition from past methods of image creation to generative AI. This study will examine the hybrid nature of AI-generated images, which blend autographic and allographic elements, analyzing their impact on visual literacy. In particular, it will focus on the ways synthography accelerates the erosion of the documental value of images thereby undercutting authenticity and reliability of visual media, and raising questions regarding artistic authorship and the artistic value of AI practices. - What comprehensive theoretical framework can be established to define and understand AIgenerated images? - How does the epistemological shift driven by TTI technologies transform the creation and perception of images? - How do TTI technologies influence the erosion of the documental value of images? - What are the implications of integrating TTI technologies into education for enhancing visual literacy? - What are the potential technological advancements, socio-cultural implications, and limitations of current TTI tools? - How do synthographies impact global visual culture? With the advent of advanced AI technologies, the process of image creation is undergoing a profound transformation. The launch of DALL·E by OpenAI in January 2021 marked a significant milestone, allowing users to create images from textual descriptions. This capability has sparked extensive debate and reflection on the nature of these new forms of imagery. This project aims to provide the theoretical and practical framework for understanding the aesthetic, educational, sociosemiotic, and technical implications of synthography, images generated through TTI. The main challenge is to explicate how AI-generated images influence traditional notions of authorship, creativity, and the documental value of images. This challenge will be addressed by taking an interdisciplinary approach, combining resources and methods from aesthetics (art theory, visual culture), education (curriculum studies, digital literacy), and engineering (AI technology, image analysis). Current studies focus on the technological capabilities of TTI. By contrast, this project broadens the investigating and refocuses on the subjective reception of AI-generated images and its social, cognitive and educational impact, thereby covering a significant lacuna. Attention to the subjective, social and cognitive implications of AI-images is expected to highlight under-explored resources for enhancing visual literacy, fostering creative expression, and developing educational frameworks that integrate AI technologies effectively.