and I grok and generally agree with the criticisms ITT.
The question that interests me:
Given some unit of time T, measured between "that video of a surreal and unstable Will Smith shoving spaghetti in his mouth", to this, is March '23 to Nov '23: call it 3.5 years.
How many units time T until the criticisms ITT are by broad consensus no longer valid, because this sort of thing has gotten that much better?
Prediction: much less than 1T. Because the systems that produced that change from then to now are non-linear.
I was at a puppet head making workshop recently, and the facilitator passed around a 3d print of a puppet head designed by an LLM. I found my eyes glazed over it, even though I found some of the hand-sculpted clay heads utterly fascinating.
Generally, I want to consume much less LLM output than I do. I'm confused why people care about projects like this.
Back in the day, there was this discussion about music. Some people prefer going to live concerts, where it's easy to get "a sense of the presence of the creator." We believed that this was an appreciation for musicianship, as opposed to aesthetics. Others prefer listening to music at home. I'd maintain that this is a purely aesthetic appreciation. And the day AI can generate genuinely good music is very near at hand, I think...
Thing is, though... Man... That "Whitepill" song is really something. The aesthetics don't trouble me. The song, which is mediocre at best, is unimportant. What troubles me is that it's more brazen than Soviet propaganda from the 1970s in its message and delivery.
If anybody believes that "universal high income" is "the most likely outcome," they really haven't been paying attention. Totally clueless.
only if you stopped reading in Aristotle.
> and, applied to humans, genetic fitness.
wtf?
> We believed that this was an appreciation for musicianship, as opposed to aesthetics.
there are other aspects to the aesthetic appreciation of a live show than purely the music (even if you simply consider that a live show has a visual element to it that can also have aesthetic attributes).
ultimately, trying to isolate aesthetic qualities both from each other and from the human experience doesn't seem like a fruitful pursuit.
> And the day AI can generate genuinely good music is very near at hand, I think...
nah
> If anybody believes that "universal high income" is "the most likely outcome," they really haven't been paying attention. Totally clueless.
we agree here
If the facilitator told you afterwards that actually, the 3d print was designed by a human and the hand-sculpted ones were made by a machine how would that change how you feel about this?
And if they again told you that they tricked you and both of them were made by hand, or both made by a machine?
I also watched the "Escape Velocity" linked on that page and it gave me a better idea why I find it hard to care about either of the "films":
> A techwear fashion show where every Look is a Silicon Valley meme, walked by an AI model who is also a fashion model. A split-flap board counts down 18 MONTHS TO ESCAPE THE PERMANENT UNDERCLASS until it flips to THERE IS NO UNDERCLASS and the catwalk lifts off.
> Blue-black night and steel blue, one red lamp as the only warm light, a colour halftone over everything, garment tags, split-flap boards and computer-vision boxes tracked to her. She wears look 00: a white cropped puff-sleeve shirt, a black pleated skirt in coated nylon with a harness belt and a hanging garment tag, black knee boots.
^ these two paragraphs fully describe what I saw in "Escape Velocity". I read these paragraphs after watching it, and they fully reflect what I saw. There was nothing more to that music video.
I can't recall any good human-made music video that you could describe in two paragraphs well enough for me to say "Yes, you fully captured it, there's nothing more to it". Even trivial details like facial expressions or prop placements add some flavour, but here I don't see anything on top.
Given that the video is actually a commentary about LLMs and AI, I can look past all that though.
Unfortunately, it seems that the majority of stuff like this is by people who don't want to learn art and just want to skip to having a thing that looks like art.
I think if you want to have good AI art you're going to have to convince artists to start use it.
Shameless plug: https://www.fathersnight.net
I wrote and "illustrated" this children's book, acting more like Warhol in The Factory [1], providing copious amounts of art and creative direction notes, building an editing platform for me to hand edit and to provide more nuanced guidance alongside edits for illustration tweaks. All prose was written by me, no LLM-assistance there.
I think we'll soon see how artists are transforming typical practices with LLMs once we get past the binary "AI art bad" phase.
Seriously, WTF is this video supposed to be? There’s no edge to it, it’s not visually very interesting, and the story just doesn’t make sense… we have a data center with red riding hood and some wolf and some singing pigs talking about sand turning into chips. What?
Which means somewhere out there, there's someone who's artistic enough that can create something abstract that would make you forget the process in which the creation was done. And make you feel the empathy, through an AI generated picture.
I've yet to see one myself, however.
This is boring, meaningless and a bit offensive.
unfortunately the only interesting thing about it is the discourse. the “art” itself sucks.
Yeah, was not hard to tell... Like seriously what am I supposed to do with this information?
> Two shots. v1 was made in one pass from a single brief and a plan sign-off (5 h 08 min). The director heard it and gave one note: wrong song. The take I had picked without hearing it was the worst of the fourteen. v2 is the whole film re-shot on the twin take they picked by ear (3 h 01 min). There were no notes on the picture in either pass.
The video is just as soulless as the video which was for me a tough watch because it just looks so off (like a lot of AI videos do).
> Claudia explains the hazards of believing she is either sentient or infallible, the myopia of AI utopianism, and how gatekeeping and the "asymmetries of fluency" can be wielded against an unsuspecting world.