vitaut 🤍❤️🤍 🇺🇦
Hacker Memes
Daniel M Karlsson@i0null [chuckles]
jan Anja 
@i0null same with programmers
Shirley Eugest@i0null This describes it well.
Dogzilla
Torsten
Chris Lamke@i0null Perfect.
bison ✅
Cattivo Strega ✅bonafide🇨🇦@i0null
I’m supposed to live on the ultimate compliment of having my works stolen,
FOR YEARS.
AI is where my work’s stolen faster by recognized corporations and sold to those who are too lazy to steal them, themselves.
People have complained because I work to ensure the digital colours are not irl.
Asiny@i0null hahaha oh jesus
PeterBSchmidt@i0null purrfect.
Maria Schnee
Grey Goo
Idealistic Pragmatist 
xs4me2
jfml - Jonas Laugs@i0null Source? Credit the artist!
@jfml oldest template i found is this one, unclear if original, welcome attribution. https://imgur.com/gallery/LDuPTk5
@i0null Thanks for looking into it!
Del C 
@jfml yeah, i wish i knew whose art this is, because it's super cute.
walter4096zero-sum interpretation.
more like the AI is taking the fish, breeding them, and putting them back in the river ..it's a tool that artists can use, even for free
Paul SchoeYour interpretation does not place much value on original creation.
It seems that you do not mind AI using artist's work as long as AI reproduces it and makes it available to the public.
I wonder how you feel when you create something beautiful and then see its value diminished because it is being replicated hundreds of times in your community, even with other people putting their name on it.
the cat is out of the bag
https://twitter.com/_akhaliq/status/1716831652545208407
this model only needs12m examples from scratch.
Disney can *easily* train inhouse without breaching IP.
You need to be thinking 'how can I use this' not 'how can i stop it' (you cant)
AK on X
Matryoshka Diffusion Models paper page: https://t.co/bJmNwaSyRH Diffusion models are the de facto approach for generating high-quality images and videos, but learning high-dimensional models remains a formidable task due to computational and optimization challenges. Existing…
@walter4096 @paulschoe it's not breaching IP if they already own the training data.
The questions about stolen surplus are legitimate, if humans are subject to copy laws, then why shouldn't machines?
this is where its ambiguous IMO,
we all carry traces of what we've all seen in our heads.
e.g. I can draw x-wing fighters, darth vader etc from memory.
but I can't sell those.
same with AI art. you can't sell images of IP
some rulings have been made like this
I think there's a compromise to be found that lets us all benefit from this new capability (ability to copy *generative rules* vs finished works)
there's more at stake than entertainment.
the world is in trouble, we have big problems we're not solving fast enough with our own minds.
These image generators are evidence of visual intelligence, which would be available as part of other problem solving processes
check this out
https://twitter.com/svlevine/status/1714307592875647291
SD 'image edits' used to help robot motion planning given a goal (imagine intermediate steps)
Sergey Levine on X
Diffusion models make great images. But can they drive robots? Usually that gets complicated really fast. We figured out how to get a Stable Diffusion model (based on Instruct pix2pix) to drive robotic instruction following. Simple recipe, works on a wide range of tasks. Thread👇
Paul H 🏴🇪🇺@paulschoe @walter4096 @i0null
I'm not an artist, and your point is very valid. I think the previous post was meaning that AI could be used as a creative tool. In the original cartoon, I imagine an artist behind the AI cat, working the AI cat like a puppet (not, by any means suggesting, of course, that AI is a gateway to plagiarism)
Please take these comments in the spirit intended.
AI art generators have limits.
I'm doing solo gamedev, I'm interested in using these generators as an assist.
But they're nowhere near replacing a real artists holistic design skills (Why would they be at 0.1p/image?).
I still need to find a real artist.
artists need to focus on [1] what AI art can't do (a lot) , and [2] seeing where they can streamline (eg do a personal finetune, do img2img).
whats coming IMO is a tool where an artist will be able to go from storyboard -> film, i.e. someone who does comics today is doing animation& films tommorrow.
it wont be big budget quality, but it will be better than hand-drawn anim
still images are indeed cheapened but demand for artwork more broadly is limitless with films,games,VR
Now how this will play out more broadly is an open question.
Remember, LLMs are doing the same for intellectual tasks, like coding, medicine etc..
we're all in the same boat
use these tools, you'll see they have big limits.
the human brain is a much bigger net,
but AI can be copied - learn once, copy to all machines instantly where each human had to learn from scratch.
which will win? things will get wild.. bring it on, IMO
vapid consumer@walter4096 I think you mean chopping the fish up, glueing them back together in different shapes and throwing them back in the river? That shit's dead and useless pal.
chopping describes a destructive act where the original fish are no longer available.
if its dead and useless, why are people posting these memes?
@walter4096 If you would like to get pedantic about the analogy, take it more towards creating clones with scrambled DNA which compete for limited resources, damaging local populations, yet which are incapable of producing successful offspring.
People make these memes to warn others of grift in an approachable manner.
To those reading along: beware those who think technology can solve social issues.
if the scrambled DNA makes them incapable surely they're no threat.
if concerned about flooding..
artists can have their own finetunes or distribute variations or turn images->anims.
so many possibilities
Felix
Sue is Writing Solarpunk 🌞🌱@i0null @Dogzilla Beautiful. And now we need a 4th frame showing Nightshade* in the bucket.
*AI poison applied to images to protect artists
https://www.technologyreview.com/2023/10/23/1082189/data-poisoning-artists-fight-generative-ai/
datensalat :firefish:@i0null thanks to #StableDiffusion hundrets of thousands of gerenative ai enthusiasts (including myself) learned that Greg Rutkowski is an amazing artist. Poor guy 😔
Kotes@i0null Exactly!
AL@i0null
this is the best image I've seen of the AI situation today. thanks
this is the best image I've seen of the AI situation today. thanks
Ryan Schultz 🇺🇦@i0null @tonroosendaal i wonder, however, if this will ironically make the artist's work more valuable. What humans value, is always a moving target... but is often couched in scarcity, it seems.
@theoryshaw @tonroosendaal yes, once they've depleted all the artists, they'll have nothing left to train on.
i think there will be a stronger market for 'AI-free' art, as well as for hybrids.
coupland@i0null You forgot to show the "Artist" cat fishing out of another "Artist" cat's bucket.
All art is influenced by other artists. You go to galleries, you peruse books, adopt techniques from other artists, go to school and read copyrighted books and analyze what makes other pieces of art great. Then you go out and create new, derivative works that are based on your body of knowledge. It's called "getting an education."
That is literally and precisely what AIs do.
Magnus Ahltorp> No, that is not what “AI” does.
"No u" Is not an answer. You're refusing to enage with the topic.
I said that AI learns by consuming a massive volume of information. Some of it copyright, some of it not. It then creates original, albeit derivative works based on its body of knowledge.
This is exactly how humans create art, build bridges, write novels, program in Python.
How is it different, aside from "I want a cut of the action when a computer does it?"
@coupland @i0null There are vast differences in what humans do and “AI” do. “AI” image generation is most similar to collage processes, but with no actual intelligence.
One very fundamental (though by no means the only) way it is different is that current “AI” models require retraining on source data before new material can be used, and has no integrated feedback.
The “intelligence” you perceive is in the mind of the operator and your own mind. It is no more intelligent than a loom.
No one is implying that lakes exist in a vacuum. Our idea of art (as with language, concepts, etc..) is socially constructed. I'm sure few would contest that.
But here the AI could take a first principles approach and fish alongside the artist. Instead it's screwing over the artist like the bourgeoisie screw the proletariat.
@i0null @coupland @velinion @oliverherold @porru
some see AI as a tool of a few big companies.
but some big companies have an incentive to opensource their models (e.g. Meta is behind OpenAI , so to counter that threat, they've released free opensrouce LLMs to reduce demand for OpenAI)
I'd like to see everyone running their own AI locally (text+images) instead of leaving it to cloud services
ironically anti-AI campaigners are going to kill this possibility if they win
@walter4096 @coupland @velinion @oliverherold @porru
If the models and datasets are open i think that does exemplify to some extent. However soon they will be so large that it will be impossible to run locally. obviously open models are concern to those who want to gatekeep but also those who want to circumvent restrictions. There are legit safety concerns about all this but I’m convinced a few in the industry would rather talks about sci-fi X-risk as a way of diverting attention away from the more prescient risks.
@i0null @coupland @velinion @oliverherold @porru
Local models have a secret weapon - finetunes which can be done on high end gaming rigs.
The bigger AI's do 'MoE' where they branch between several experts.
Opensource models can in effect be the ultimate 'MoE' , distributed across the web.
IMO open models are the way to ensure we all have maximum input inguiding how they're used, what they are, and ensuring widest benefit
@i0null @walter4096 @velinion @oliverherold @porru The kind of prescient risks we should be talking about wrt AI are things like job displacement and wealth disparity. AI will create trillions in wealth how do we prevent those trillions from going to, like 2 people. But that's an economic & ethics discussion, it really has nothing to do with AI itself.
Instead we're distracted by "how do I get paid for every ChatGPT search" or "AI=evil" because it drives "likes" or "ZOMG! Extinction event!"
@coupland @i0null @velinion @oliverherold @porru
yeah this is why I back opensourced AI. I'm resisting using GPT and giving mindshare to the less capable LLaMa's that you can run locally.
people kind of sleepwalk into this centralization of power by becoming *too* dependent on cloud services. the good outcome is where we all keep buying powerful PC's and do as much locally as possible
you could be paying for a big tech service (->disparity) or it could be saving you money (lift all boats)
@walter4096 @i0null @velinion @oliverherold @porru
> people kind of sleepwalk into this centralization of power by becoming *too* dependent on cloud services
Just like how we've sleep-walked into surveillance capitalism and climate crisis. Why tackle tough issues when we can be lulled to sleep by shiny toys?
Really appreciate all the perspectives, great discussion.
@coupland @walter4096 @velinion @oliverherold @porru
Exactly thats what I was refereing to. Accellerationism is going to exacerbate inequality. That trillions in wealth isn't just created, it's the potential to siphon it from other parts of the economy (mainly labour) and one of the problems is it's going to be a slow burn.
As in rather then AI strait up replacing people, it will simply slow new hires. And how can an economy address that without some kind of global body? A global body could also be a gift to the top AI firms that are able to lobby for regulatory capture.
@i0null @coupland @velinion @oliverherold @porru
yeah regulatory capture is exactly my fear.
I'm just a lone tinkerer at the minute - I see huge potential being unlocked with these opensourced models , and a bunch of people arguing they should be shut down :/
I've been pro-tech all my life ever since my first 8bit machine. I've always seen it as an individual enabler, and without being 'left or right wing', opensource is very inline with that.
@walter4096 @coupland @velinion @oliverherold @porru
Yeah it's such a shame that a technology with such emancipatory potential is more likely to hand greater power to the biggest players.
Many people say technology is neutral but the application of it certainly isn't.
I’ve been fairly surprized by Meta's commitment to open source but after that whole Cambridge Analytica scandal IMHO they're going to need to do a whole lot more to regain some reputation.
@i0null @coupland @velinion @oliverherold @porru
it's great how this emerged as a balancing mechanism
meta, google, microsoft,apple all saw what OpenAI were doing, and panicked.
MS bought OpenAI to stop them growing as the next tech giant(ie conversational interface)
Meta then opensourced their own (inferior) LLMs ("ok, we dont think we can get to 1st place, lets at least empower opensource community to stop Microsoft dominating completely")
Velinion@i0null
This implies artists don't learn from other artists, and that AI aren't trained on any data not produced by artists. They should both have a fishing line in the bucket, and in the lake.
This implies artists don't learn from other artists, and that AI aren't trained on any data not produced by artists. They should both have a fishing line in the bucket, and in the lake.
Theodore John Kaczynski@i0null Welcome to open source