Hacker News Reader

Comment by kengor | Hacker News Reader

Not trying to denigrate the amazing technical skill needed to make this thing, but from the point of view of someone who makes art, this is a pastiche image maker. I have yet to see AI generate anything that's not a combination of existing images and techniques. Probably wading into murky philosophical waters here but I reckon art needs to be more than a nice 'arty' looking image to be considered any good. If you are going to make Art that is worth the capital 'A' on the front, then it has to mean something. However I suspect the days of actually making Art are in the past. We're on to something else now. Not necessarily better or worse, just different. Art is an old idea, not entirely relevant any more.

View on HN

Not trying to denigrate artists, but as somebody that admittedly doesn't appreciate art, I'm still not sure how artists aren't just pastiche image maker themselves. Based on your comment, I'm not entirely clear what the difference is

Can this machine, beautiful art though it may make, look at the work of the cubists and spontaneously generate Gehry?

Without Mondrian, would it find beauty in clean abstract lines and sharp angles? Can it, by itself, come up with the concept of neoplasticism and implement that, in any way, to create a new beautiful artwork?

Would the AI, not for Klimt, have an appreciation for how casually erotic the female form can be?

Does it, yet, have this appreciation?

I don't know, and I'm sure it doesn't either.

> I don't know, and I'm sure it doesn't either.

Even more pertinently, it doesn't care.

I might stretch it further that it can’t even care because the machine does not have the capability to care.

I think to the average human, that was implicit in my comment, but for the benefit of the machines, it's good to spell it out.

Well said. I actually do appreciate many forms of art, but I feel that many artist (that I know) are just a mixture of regurgitators and expert salesmen. They make something random, then come up with a backstory that makes it somewhat interesting. This AI can out perform them, without the need for the ingenuous hook story.

The serious ones who stick around doing the same thing for a lifetime do eventually find some meaningful essence however abstract that is. But there is indeed a lot of deception in the art world since it is a non quantifiable subject the story does add some value

> I actually do appreciate many forms of art, but I feel that many artist (that I know) are just a mixture of regurgitators and expert salesmen. They make something random, then come up with a backstory that makes it somewhat interesting.

I wouldn't call your relationship with art "appreciation". Sounds more like undisguised disdain.

I appreciate art, not artists. You seem to either have misread what I said, or are actively trying to misattribute my words.

> "I appreciate art, not artists."

  Can you have one without the other ?

  Apparently, now, you can...

Interpreting the random is a fair form of art. Like a sports commentator or video game mechanic.

https://youtu.be/9PKp4q5-Aww?t=289

Watch this video of an art dealer looking at the "efforts" of the Top Gear trio - He's looking for premeditation, emotion, and inspiration, it's arguably a bit pretentious, but you can see his change in the thought process when James May starts to explain how his big metal face is a deconstruction of a car into a display of his emotions while driving it.

A piece of art is different to different people, but you are consuming the artist's opinions both directly (if you are aware of the artist's previous work, i.e. Francis Bacons paintings are pretty interesting if you know a bit about his life) and indirectly (The Video game Pathologic is very polarising but is so utterly alien to the archetypal video game - especially when it first came out - that it really resonates with some of those who play it). Imitation and technique only go so far - Erik Satie wasn't particularly well regarded as a composer (or even a musician, he often referred to himself as a Phonometrician) when he produced his most famous works but he is known to this day for his command of minimalism, repitition and a surprisingly Jazz-like use of "horizontal" harmony from simple scales (Potentially constructed by stacking triads but I don't know)

(pastiche: I had to look it up)

    pastiche

  n 1: a musical composition consisting of a series of songs or other musical pieces from various sources [syn: medley, potpourri, pastiche]

  2: a work of art that imitates the style of some previous work

http://www.dict.org/bin/Dict?Form=Dict2&Database=*&Query=pas...

A lot of art is exactly that, but it's usually very obvious. As it is in these pictures.

I find this whole discussion quite interesting: There seem to be many people who think these photos are pretty much just like what you see in an art gallery, and others who say that it's obviously just inspirationless imitation.

Comparing this to generated music -- every time there's a "quiz" challenging to distinguish computer-generated snippets of music from real classical music, I've gotten 100%. The key difference is that the computer-generated music I've heard so far is just one musical idea after the other; it doesn't seem to have intent or motion. (Much like GPT-2 style text: a single sentence looks reasonable; once you get to several paragraphs, it completely stops "hanging together".)

> it doesn't seem to have intent or motion

Because it does not have any. It’s a tool. Can we say a cool brush creates paintings, a complex instrument creates music? By itself Im afraid it’s far from it. A human hand must be involved somewhere (im literal about the hand part, but a brain nonetheless)

I agree with you. I think the reason AI hasn't made anything but derivatives is in part because machines are not sentient. 100% of AI is machine learning, there is no heart or conscience.

However art is anything that moves you and the paintings on this website are most definitely moving to me. I find them terrifying and a lot of that has to do with the fact that a machine made them.

There isn't anything really grotesque there. (Grotesque is a blend of horror and empathy. Horror is a scary guy in a mask holding a bloody axe or things jumping out at you. Grotesque is a person maimed.)

This site has, in my opinion, art that is nauseatingly scary. It is scary in part because its soulless and part because it's a mirror of how a machine sees humanity.

Regardless of how it was made, it moved me therefore from my perspective: It is art.

There really is something nauseating or repulsive about this. I've had the same feeling with other ML generated images. It's alien and familiar at the same time.

It reminds me of this nightmare that was making the rounds a while back: https://sea.mashable.com/culture/3430/people-cant-figure-out...

It is only familiar because it was trained on familiar work.

> 100% of AI is machine learning, there is no heart or conscience.

And, because AI implies heart and/or conscience, ML is a much better term because it is more accurate and describing.

Very well stated

I’ve been exploring this space and find myself thinking of Andy Warhol. I would definitely consider him an artist. A great one even. But his work was quite different from Picasso. Warhol had a factory. His silkscreens were often done by assistants. It said something about the Modern Industrial Age. His work raised questions about authorship, authenticity, etc.

I found myself thinking about similar things when I trained a neural net to make art: who’s the artist? Me, or the machine?

When did art ever require purpose? Who decides what is and isn't art? Very tricky path to trod down.

I thought one of the ideas was to build into this a way to let people pick the good stuff.

Which is to ask if Audiences are never wrong?

it isn’t possible to create something that doesn’t serve a purpose. what “good” artists and other art professionals have that makes them experts is 1) an understanding of this fact, then 2) knowledge of semiotics and history among other related areas that afford as deep an understanding of potential meanings as possible — this is the tricky part, because it requires real work

In this case, you decide. You click that Favorite button and it's a decision you make. The AI artist has a set of rules that learned from the previous trained art pieces, but the final shape, and the final decision is yours. You like it or not, and it's quite personal too.

What if I make the argument I think this art is more beautiful, because of the interesting way it was generated?

Relevant quote from Feynman about art: https://www.goodreads.com/quotes/184384-i-have-a-friend-who-...

I like art too and make some myself (musical). But I happen to think art is just what each viewer thinks of it. I don't know enough about visual art to know whether this is a "pastiche image maker", though I suspect AI won't make that for long, if it is.

All art (and innovation, really) is about combining existing things in new and unexpected ways. Doesn’t matter whether it’s a machine or person that creates the end result, the art is in the process.

It’s just that with machines and deep learning we can try more variations in a much shorter period of time.

No more waiting to see whether that bunch of disaffected art students sitting around drinking absinthe and smoking strong cigarettes come up with anything good, get a bunch of computers to do it in a fraction of the time - and with far less angst.

Those computers don’t really think. Whatch what you wish for, future might just look like that.

That art moves person who sees it is more important than how original it is.

Originality is how artists signal to their peers and scene their value. It's relatively modern concept. Art existed before originality was considered a value and it probably exists after the originality stops being an issue.

I have yet to see 99.999% of artists "generate anything that's not a combination of existing images and techniques". As mere mortals are not in a position to hang the works of 0,0001% that can this AI art can be a reasonable alternative.

I wonder, based on your comment, what the output would look like if the training was inverted. Anything that looks like art would skew towards zero, for instance.

That would be intersting

Reply on news.ycombinator.com