Citat fra
This Is the Project of a More Just World’: Trevor Paglen on Making Art That Shows Alternative Realities

Actually, when you look under the hood at how AI is actually done,
it requires tremendous amounts of human labor.
To create those images you need massive training sets. In my case, I need
hundreds, or even thousands, of images to create a training set that I use to make one of these pieces.
You assemble those images and then label them and tag them in order to teach the AI what these images are.

Citater fra
The AI Art at Christie’s Is Not What You Think:

Machines are not making art on their own.
Most would say that AI is augmenting artists at the moment and the description in the news is greatly exaggerated.
A human chose the data set.
A human designed the network.
A human trained the network.
A human curated the resulting outputs.

Paglen’s are neural networks with deliberately complicated data sets,
visualizing the human opinions embedded in their training.
Instead of showing how independent and advanced GANs are,
as Obvious’s “purist” work purports to do,
Paglen’s work shows that GANs are dependent, limited, and hugely vulnerable to perpetuating the hidden biases held by the developers and researchers guiding them.

Agerbos Side