Beyond Precedents: A Cross-Disciplinary Approach to Text-to-Image AI

Following the surge of publicly available artificial intelligence platforms like Midjourney and Leonardo. Ai in 2022, text-to-im-age AI has been a valuable tool for idea generation in the early stages of the design process. In the classroom, the potential of AI allows design students to craft custom reference images and concept variations, utilizing large datasets without the need for advanced technical or language skills. When used for reference making, this raises the question: can design be created solely on precedents represented in large datasets? This project’s primary objective is to contribute to the archival of early text-to-image explorations by organizing and categorizing AI-generated images derived from resident-driven text prompts. Additionally, the project aims to explore whether design can be created solely on precedents, focusing on the critical examination of AI-generated images and their representation of marginalized communities. The overarching goal is to improve student work, leading to more meaningful, culturally-informed precedents and appropriate design solutions.

https://doi.org/10.35483/ACSA.AM.112.81

Volume Editors
Germane Barnes & Blair Satterfield

ISBN
978-1-944214-45-6