The AI Brain in the Cultural Archive
A shorter version of this essay published in MoMA magazine, July 21, 2023.
Another earlier version was included in Chapter 5 of the “Artificial Aesthetics” book.
The later reworked version became part of ‘Make it New’: GenAI, Modernism, and Database Art" article (2024).
From the essay:
"On the surface, the logic of modernism appears to be diametrically opposed to the process of training generative AI systems. Modern artists desired to depart from classical art and its defining characteristics, such as visual symmetry, hierarchical compositions, and narrative content. In other words, their art was founded on a fundamental rejection of everything that had come before it (at least in theory, as expressed in their manifestos). Neural networks are trained in the opposite manner, by learning from historical culture and art created up to now. A neural network is analogous to a very conservative artist studying in the “meta” “museum without walls” that houses historical art.
But we all know that art theory and art practice are not the same thing. Modern artists did not completely reject the past and everything that came before them. Instead, modern art developed by reinterpreting and copying images and forms from old art traditions, such as Japanese prints (Vincent van Gogh), African sculpture (Picasso), and Russian icons (Malevich).
When it comes to artistic AI, we should not be blinded by how these systems are trained. Yes, artificial neural networks are trained on previously created human artifacts. However, their newly generated outputs are not mechanical replicas or simulations of what has already been created. In my opinion, these are genuinely new cultural artifacts with previously unseen content, aesthetics, or styles."