Curated by Artificial Intelligence: AI-Human Collaboration in the Art Museum
Document Type
Article
Publication Date
5-14-2024
Abstract
In 2023 the Nasher Museum of Art at Duke University embarked on an experiment to use artificial intelligence to curate an exhibition from the museum’s collection. This was, to our knowledge, the first time a museum had done so. Using tools provided by OpenAI and several other data science frameworks, we transformed publicly-accessible information on our collection into statistical representations of “relatedness.” This dataset included fields such as artist, title, date, medium, cultural group, keywords, and description for the nearly 14,000 objects in the Nasher’s collection. This custom version of ChatGPT was thus “educated” about the Nasher’s collection and able to access, select and make connections among specific works. Our team further developed a series of prompts and instructions for ChatGPT that asked it to select artworks for the exhibition. We followed a similar process in the creation of the accompanying wall and label texts and the gallery's design. The resulting exhibition, Dreams of Tomorrow: Utopian and Dystopian Visions, curated by AI, featured 21 artworks from the Nasher’s collection and was on display in our Incubator gallery from September 9, 2023, to February 18, 2024.
This presentation explores the capabilities and limitations of AI as curator, the use of the exhibition as a pedagogical tool, and the reception of the project by our university and community visitors. We will also share our technical process for creating a custom ChatGPT interface, as well as recent advances in this technology that would have provided easier ways to enlist AI as a curatorial counterpart. While museum professionals are far from relinquishing control of exhibition making and interpretation, this exercise was a powerful way to explore the applications of AI in the creative realm as related to curatorial authorship and expertise, the subjectivity of the selection process, and the future impact of technology on museums. We ultimately discovered that although AI was not a substitute for the human curator, it could be a powerful curatorial collaborator.
Recommended Citation
McHugh, Julia Ph.D; Olson, Mark Ph.D.; and Price, Marshall Ph.D., "Curated by Artificial Intelligence: AI-Human Collaboration in the Art Museum" (2024). N.C. A&T Faculty Enrichment Workshop - Aligning AI Themes of Future Teaching, Research, and Interdisciplinary Collaboration. 5.
https://digital.library.ncat.edu/facenrichwrkshopai/5