Christiane Paul version 1.0
00:00 Jon Ippolito introduces the Digital Curation program and Christiane Paul
01:31 Will AI lead to an "artificial retirement" of artists?
04:00 Unlike AI-generated imagery, AI art requires critical engagement
We are now in the third peak of AI hype.
08:07 1960s: Cohen's motivation and algorithms.
13:48 Physical plotters live in the gallery.
18:09 Copyright for an AI "expert system."
19:22 1980s: artists begin to explore machine vision, AI agents, and custom datasets
Artists include David Rokeby, Lynn Hershman Leeson, Stephanie Dinkins, and Memo Akten.
24:05 2010s: GANs (Generative Adversarial Networks)
AI-generated portrait by Robbie Barrat/Obvious sells for $400k at Christies.
Artists include Mary Flanagan, Devin Kenny, and Beth Coleman.
Writers include K Allado-McDowell and Mark Amerika. Ubermorgan creates an entire exhibitions from scratch via AI.
"Most writers working with these systems use earlier versions because GPT-4 is so neutered in its language."
29:48 2021: Text-to-image models
Artists include Holly Herndon and Mat Dryhurst, Boris Eldagsen, Bennett Miller, and David Salle
"This is becoming a nightmare for conservation...all the works that are working with text-to-image systems are locked into large corporate systems where the question becomes how can we even preserve this other than documentation and snapshots"
An image [made with generative AI] should not be in a photograph contest because it has nothing to do with photography, there is no photography involved in this process; it is an AI system imitating photography....a nostalgia for something that never existed"
35:06 Conforming to the technology versus influencing its evolution.
35:33 How does an "expert system" work if it doesn't rely on training data? (Matt LeVan)
38:41 How relevant is the unseen process for AI (and by extension digital) art? (Michael Katz)
41:26 Do AI artists have to sacrifice control over the aesthetics of their work? (Yan Chmarau)
45:04 What does AI do for a painter like David Salle other than speeding production? (Michael Katz)
47:30 Can artists find sustainable ways to work with AI in an era of climate change? (Nimrod Astarhan)
50:05 How have artists tackled the problem of AI's "black box"? (Vienna Kim)
52:05 How did Cohen resolve the problem of his machines upstaging his software? (Eryk Salvaggio)
54:40 Would AGI (Artificial General Intelligence) change attribution for posthumously generated works? (Darian Pena)
60:07 What is the best strategy for preserving AI art? (Tanja Kunz)
63:18 For emulated works, how do you adjust for the increased speed of newer machines? (Chris Coleman)
64:57 Does AI's dependence on language preclude it from purely visual expression? (Michael Katz)
This teleconference is a project of the University of Maine's Digital Curation program. For more information, contact ude.eniam@otiloppij.
Timecodes are in minutes: seconds
How are traditional expectations of authorship, copyright, and provenance buckling under stress from AI?
On 16 April 2024 UMaine's Digital Curation graduate program organized a conversation on the interplay between art and AI with Christiane Paul, commonly regarded as the world's foremost digital curator.
As Curator of Digital Art at New York's Whitney Museum, Christiane Paul has commissioned over 100 projects and organized dozens of exhibitions from Madrid to Moscow.
The winner of last year's International Media Art Histories Award and numerous other accolades, she has also taught at The New School, School of Visual Arts, Rhode Island School of Design, and San Francisco Art Institute. The author or editor of eight books, Paul opened this spring a provocative exhibition on groundbreaking AI artist Harold Cohen.
Watch the entire video or choose an excerpt from the menu on this page.
Or view more teleconferences from the Digital Curation program.