Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AI doppelgänger experiment – Part 1: The training

julienposture.substack.com

Post reply on HN
Print view View original

11–20 of 137 posts

Re: AI doppelgänger experiment – Part 1: The training

#11

Post by ganzuul »

jsheard wrote: post #8
ecjhdnc2025 wrote: post #5

Earlier quoted context omitted.

> The art masters cannot be replaced, you cannot predict what they’re going to do next because they change the rules entirely. This is how we know a great artist from a mediocre artist. I agree that in principle the best working artists are those who know how to create, modulate and discard artistic languages over time -- creating and abandoning rules for sequences of work -- but AI systems can clearly be unethically…

> because as soon as there's enough work in the new style it can be stolen. This is where the oft-cited comparison to previous advances like the camera breaks down - AI has an ongoing parasitic relationship with the mediums it is displacing. The camera doesn't need a thousand photorealistic paintings of a new subject to be made before it can capture that subject going forward. AI is an unprecedented situation, as far…

This week I completed a sculpture which I intentionally made to be non-photographable by choosing lighting and materials which showed off high reflectivity and high chroma so that screens would not be able to reproduce it. To the human senses it is something that really steals your attention, perhaps because we have becomes so used to the constraints of our reproduction technology and this is not that. The Prussian blue pigment sucks you into the astral realm and the gold is hypermodern reference to the Apollo program. It started a lot of conversations.

The work is a critique of photography, prints and screens and all of postmodernism. It also is also a warning about being satisfied with mediocrity.

Re: AI doppelgänger experiment – Part 1: The training

#12

Post by spywaregorilla »

voiceblue wrote: post #4

Related: "Invasive Diffusion: How one unwilling illustrator found herself turned into an AI model" [0]. It's astonishingly simple to train both LoRAs and Embeddings nowadays -- if you combine it with tech like ControlNet, you can more or less have direct control over the illustration as well (see this example I converted to Hollie Megert's style with barely any effort [1] using the model from [0]) -- but FWIW I was p…

> However, I do think there needs to be some kind of protection from having your 'style' commoditized without consent, using unauthorized images of your own art.

1) Hire artist to copy the style for 10 images

2) Train model on those

problem solved.

If you choose "using your images specifically" to steal your style as the hill you want to die on, then you're going to find it's a very useless line of defense.

Re: AI doppelgänger experiment – Part 1: The training

#15

Post by nicklecompte »

gedy wrote: post #7

The challenge I see is that (in the US at least), "style" is not copyrightable in the arts. IMHO this is a good thing because artists learn and copy from one another, and if it were copyrightable that would lead to no end of lawsuits between media companies. So my concern is trying to apply copyright to training AI models can easily slippery slope into more draconian copyright rules for human artists.

I don't think there is a slippery slope in the foreseeable future, unless you buy into sci-fi views of AI being like a human. We need to update the laws around copyright in response to these machines. It's similar to why copyright laws exist in the first place: the concept was developed in response to the printing press.

At some point there will be a real I, Robot problem about an AI artist that actually understands what its drawing and doesn't depend on interpolated plagiarism of inhuman amounts of data. But we aren't even close to that yet.

Re: AI doppelgänger experiment – Part 1: The training

#17

Post by ninetyninenine »

AI is to the point where it can create it's own style. We talk about these styles as if it's some kind of sacred thing. What you don't realize is that we're past this point.

There's a latent space of different art within a certain style category, but there's the latent space of different categories itself. AI can already traverse the space between art in a single style category AND it can traverse the gradient space BETWEEN categories. It can even go beyond that.

AI has turned everything into a freaking best fit curve.

Copywrite is such a mundane question. It's just a fight against the inevitable truth. That human creativity are just points on that curve. Nothing more, nothing less.

I view this as a BAD thing. Copywrite is good, it's great. It's a law that implies what humans create has meaning. But we have to face the truth. That the output we produce is in actuality just one part of a high level pattern. That's all we are.

Re: AI doppelgänger experiment – Part 1: The training

#18

Post by vunderba »

voiceblue wrote: post #4

Related: "Invasive Diffusion: How one unwilling illustrator found herself turned into an AI model" [0]. It's astonishingly simple to train both LoRAs and Embeddings nowadays -- if you combine it with tech like ControlNet, you can more or less have direct control over the illustration as well (see this example I converted to Hollie Megert's style with barely any effort [1] using the model from [0]) -- but FWIW I was p…

Corporate Memphis seems to have that very "flat almost clip-art" style to it.

Generative art is ideal for this type of stock art use case where 90% is simply good enough for the vast majority of situations.

I feel like the gig economy (things like Fiverr) are unfortunately going to get absolutely crushed by generative art systems since most of their customers aren't that particular and are likely fine with anything that DALL-E / SD can pump out.

Re: AI doppelgänger experiment – Part 1: The training

#19

Post by LettuceSand12 »

voiceblue wrote: post #4

Related: "Invasive Diffusion: How one unwilling illustrator found herself turned into an AI model" [0]. It's astonishingly simple to train both LoRAs and Embeddings nowadays -- if you combine it with tech like ControlNet, you can more or less have direct control over the illustration as well (see this example I converted to Hollie Megert's style with barely any effort [1] using the model from [0]) -- but FWIW I was p…

Where can I learn these techniques?

Re: AI doppelgänger experiment – Part 1: The training

#20

Post by monadINtop »

ninetyninenine wrote: post #17

AI is to the point where it can create it's own style. We talk about these styles as if it's some kind of sacred thing. What you don't realize is that we're past this point. There's a latent space of different art within a certain style category, but there's the latent space of different categories itself. AI can already traverse the space between art in a single style category AND it can traverse the gradient space…

lol sometimes I wonder if the reason so many tech bros make so many grandiose statements about how AI has already trancended the bounds of human creativity is because they've never really had actual experience with art and literature and music growing up. They say things like this and then you go and look for an example of these incredible works of artificial art and it is the biggest pile of shit you have ever seen.

Just like how they think that these glorified typing assistants function identically to the human brain. I mean maybe they really do replicate the level of creativity and intelligence of the people who believe these things, but maybe the people with decades of research in linguistics and neuroscience are right to be skeptical.

Its kinda like when people talk about research mathematics like its all adding really big numbers together or calculating really tricky integrals.

Post reply on HN