4 ms·

I think AI is currently able to generate images I find pleasant to look at, but they have no impact on me beyond that momentary enjoyment. It is still decoratio

by lokl 6y ago

I think AI is currently able to generate images I find pleasant to look at, but they have no impact on me beyond that momentary enjoyment. It is still decoration and not yet art, according to my personal definitions. When I can tell an AI to make an image about the Spanish Civil War and it produces Guernica, then I will be impressed in the way human artists impress me. And then we will have a powerful new tool to communicate. I would like my own personal Guernica each day to help me learn about something happening in the world.

mikepurvis 6y ago
But how much of that is context? Is the absence of deeper impact because you're aware that a soulless NN produced these, or is it something intrinsic in them? I'm not at all an art expert, but I definitely looked at several of them and could easily picture a critic describing the significance of the relationship between the shapes and colours.
- lokl 6y ago
  I feel the same way about many human-produced works, regardless of critics' views, so the fact that I'm aware these are produced by NN is probably not the only explanation.
  - mikepurvis 6y ago
    Yeah, it sounds like in that case your criticism may be more about abstract art as a style than it is about whether a given piece was produced by a human or a machine.
    - lokl 6y ago
      Not all abstract art. Some Kandinsky, for example, is more than decoration to me.
jjk166 6y ago
"Well you asked me to paint an authentic Guernica so of course I had to bomb Guernica"
jameslk 6y ago
This doesn't sound like it would be unrealistic. I think this can be partially solved by two systems: one that paints and one that finds the best painting to match a current event (in the form of a topic). The latter system could be trained on a set of paintings and corresponding topics or meanings, and then uses this training to match newly generated paintings to topics/meanings.
- dropofwill 6y ago
  There are definitely people working on this sort of thing. I recently read this conf paper on generating modified flags based on a topics (e.g. Cyprus + peace): https://cdv.dei.uc.pt/ever-changing-flags/ https://cdv.dei.uc.pt/ever-changing-flags/ Maybe not as technically exciting, as a super generic NN, but I think projects scoped like this are useful to think about.
- wimpypistol 6y ago
  Correct. This is actually quite doable with word embedding or image classification of a topic, semantic analysis, and then providing those as input to a network like the one I used for 1SecondPainting. I love the future.
staycoolboy 6y ago
> It is still decoration and not yet art, according to my personal definitions. This isn't a personal definition. There literally is a distinction in fine arts study.

by lokl 6y ago

mikepurvis 6y ago

But how much of that is context? Is the absence of deeper impact because you're aware that a soulless NN produced these, or is it something intrinsic in them? I'm not at all an art expert, but I definitely looked at several of them and could easily picture a critic describing the significance of the relationship between the shapes and colours.

lokl 6y ago
I feel the same way about many human-produced works, regardless of critics' views, so the fact that I'm aware these are produced by NN is probably not the only explanation.
- mikepurvis 6y ago
  Yeah, it sounds like in that case your criticism may be more about abstract art as a style than it is about whether a given piece was produced by a human or a machine.
  - lokl 6y ago
    Not all abstract art. Some Kandinsky, for example, is more than decoration to me.

jjk166 6y ago

"Well you asked me to paint an authentic Guernica so of course I had to bomb Guernica"

jameslk 6y ago

This doesn't sound like it would be unrealistic. I think this can be partially solved by two systems: one that paints and one that finds the best painting to match a current event (in the form of a topic). The latter system could be trained on a set of paintings and corresponding topics or meanings, and then uses this training to match newly generated paintings to topics/meanings.

dropofwill 6y ago
There are definitely people working on this sort of thing. I recently read this conf paper on generating modified flags based on a topics (e.g. Cyprus + peace): https://cdv.dei.uc.pt/ever-changing-flags/ https://cdv.dei.uc.pt/ever-changing-flags/ Maybe not as technically exciting, as a super generic NN, but I think projects scoped like this are useful to think about.
wimpypistol 6y ago
Correct. This is actually quite doable with word embedding or image classification of a topic, semantic analysis, and then providing those as input to a network like the one I used for 1SecondPainting. I love the future.

staycoolboy 6y ago

> It is still decoration and not yet art, according to my personal definitions. This isn't a personal definition. There literally is a distinction in fine arts study.