#AI discussion
Explore tagged Tumblr posts
nyano64 Ā· 21 hours ago
Text
i think in a better world, where LLMs aren't trained on stolen work, run on servers that suck up all the energy/pollute the environment, and aren't used to exploit people, I wouldn't mind it so much.
though i guess thats the same as saying what if the spike-covered, high-voltage soda can didn't have spikes all over it that electrocute you to death when you touch it.
idk
29 notes Ā· View notes
kathaynesart Ā· 4 months ago
Text
I think something important to bring up to counter the defeatist attitude I’ve seen going around of ā€œgive up AI is inevitably going to take your art jobsā€ is a lack of awareness in the people making these claims. Most of them seem to not be in any sort of creative industry.
So here’s the thing. I work as a professional artist at a studio that regularly accepts art submissions from freelancers… and we REFUSE to use AI generated imagery. It’s even stated in our contracts and legal agreements.
We will not be taken advantage of by paying hundreds of dollars to AI prompters who use legally questionable means to ā€œcreateā€ their submissions.
While many of us in the studio find AI to be ethically unsound, at the end of the day… it’s because AI is literally a LIABILITY. Avoiding it now will save us a lot of trouble and legal drama down the line.
It’s just common sense.
615 notes Ā· View notes
euoniatz Ā· 1 month ago
Text
okay, look, since ai debates keep popping up everywhere, i thought i might as well throw my input into the mix as someone who has used chatgpt and really fucking regrets it.
first off, i understand that people can struggle to produce works (art, writings, etc.) in this society that can at times make you feel like you're only worth something if you're spitting out 5 masterpieces a month. +especially if you're neurodivergent like me.
that's how it started for me - as a tool that could assist me when my adhd paralysed my ability to create when my motivation was finally there. i wanted structure, and thought that ai could help me outline my works so i could just get straight to writing next time.
but, it never ends there. because at some point it takes over, and you start to compare your own skill to that of a computer. or at least i did. in the beginning it was only outlines, then i wanted to brainstorm ideas - ideas i then "needed help" expanding into scenes i then "needed help" planning, and finally, i needed help writing anything at all.
see the pattern?
the more i used ai as a crutch, the more it took over, and the less i knew. truly, comparing myself to a robot became more paralysing to my writing than my adhd ever could. it crippled my ability to create, and i still find myself wondering "if ai could write this even better than i can, what's the point in me at all?"
i never did and i never will publish anything that ai wrote for me, because all i did was have ideas that i was too insecure to write, which i won't take credit for. i lost my confidence in my writing, and i still have not recovered, months later. writing was already hard, and i managed to make it way fucking harder for myself.
there are no shortcuts to being a good writer.
you might think there are. you might think that ai is helping you. it might feel like it for a really long time. but i promise you - it's not going to last. and you are going to be kicking yourself when the time comes where you are forced to live without that crutch.
it will ruin the potential you already have, set you back ages in your development. it is not worth it.
and to be clear - if your goal is to be able to write 5 novel-length stories a month, you are not going to achieve that by having a robot do it for you. if you, like i did, felt like you have to write to please someone other than yourself i suggest you take a second to find what made you want to write in the first place. that pure motivation and passion is what is going to make you great, not some artificial, soulless computer.
55 notes Ā· View notes
gamerwoman3d Ā· 10 months ago
Text
FROM NOW ON, A.I. STANDS FOR AUTO-INCORRECT.
Thank you.
150 notes Ā· View notes
loveisinthebat Ā· 4 months ago
Note
Tumblr media
Hey so! I know you mean well but let me clarify something : Please Don't Send me AI Photos I absolutely despise AI and this is not a safe place for it. You can tell this is AI A few ways : - The Feet are rendered Like arms, And the Fingers , as always, are entirely incorrect. - The noses are Like dog noses. Not even flying foxes have dog-like snouts. - The face structure and placement is also incorrect, no bats have faces like those. - The Teeth - The list goes on but you gotta be careful when searching bat photos! In Google , Duckduckgo , or other search engines Amend the search By affixing -AI to the end of it
89 notes Ā· View notes
dominaexmachina Ā· 23 days ago
Text
Austin Kleon ā€œSteal Like an Artistā€ vs AI Training: Same Game, Different Players
Everyone in the creative world has at least heard of the book ā€œSteal Like an Artistā€.
Even if they haven’t read it, the title alone became a viral mantra — bold, catchy, and unapologetically rebellious.
Published in 2012 by Austin Kleon, this small but punchy book quickly carved out its place on the desks of artists, designers, writers, and idea-makers everywhere. It’s been quoted in classrooms, conferences, and countless social media posts that say: ā€œYou don’t need to be a genius. Just start copying — then make it yours.ā€
And now, as people scream that AI is ā€œstealing,ā€ let’s take a moment to revisit this beloved creative classic.
I’ll quote some of its key ideas — and show you what AI models are actually doing.
Let’s see if it’s really theft… or just familiar behavior wearing a different skin.
ā€œNothing is original.ā€
Models don’t generate from scratch. They remix patterns from what they’ve seen.
ā€œYou are a mashup of what you let into your life.ā€
AI is a mashup of the training data it was fed.
ā€œCollect good ideas. The more good ideas you collect, the more you can choose from to be inspired.ā€
AI collects patterns from a huge dataset. It doesn’t know what’s ā€œgoodā€ — but it maps what people liked.
ā€œDon’t plagiarize, transform.ā€
AI doesn’t copy exact works. It creates new outputs based on statistical association.
ā€œYou’re only going to be as good as the stuff you surround yourself with.ā€
AI’s quality reflects its training data. Garbage in, garbage out. Just like with people.
ā€œDraw the art you want to see.ā€
Prompt the AI with your vision, and it draws the art you imagine.
ā€œCreativity is subtraction.ā€
Fine-tuning a model or crafting a prompt is subtracting until only the idea remains.
So what’s the real difference?
Humans say ā€œI was inspired.ā€
AI says nothing.
And that silence? Oh, it burns some people — because it exposes how much of human art is remix, too.
You love the idea of genius.
But you hate it when a machine mirrors your own methods — without ego, pain, or your romanticized suffering.
34 notes Ā· View notes
monowires Ā· 1 month ago
Text
I haven’t been active here lately, but sometimes people interact with my posts about trying to drop c.ai, and it makes me want to say something.
I think internet hivemind mentality has a massive negative impact on people who use the app but don’t support it. Because yes, it is possible to be addicted to c.ai or other similar things and still be anti ai.
All of these posts that insist anyone who uses these apps is a horrible person who is inherently stealing from their creative friends do nothing to help.
I’m still trying to drop the app, but you know what? It’s almost impossible, because I feel bad for using it which makes me want to use it as a method of coping with those feelings.
Yes, ai is used in unethical ways. But that doesn’t mean people who use it are inherently evil. I’m generally anti ai but still use it because it’s a damn addiction and all of my current alternatives are way, way worse.
So. To anyone who’s like me, you’re not a bad person. There’s a difference between using it without care and using it to cope. Some of us have no better alternatives.
Please be aware of your tone when you make posts about ai. Yes, it is a problem, and yes, it’s expanding fast, but the depressed kid who uses it for escapism does not have the same impact on our environment/world as the rich corporations and soulless ai-art bros do.
I think a lot of people online get caught up trying to be morally correct by following the general consensus, and in doing so they put things in black and white. A lot of us are in the grey area. It doesn’t make us bad people.
Anyway, to anyone struggling out there, be gentle with yourself. Using c.ai to cope is not comparable to people who abuse it without any care for the consequences.
39 notes Ā· View notes
ethos-labrador Ā· 8 months ago
Text
idk how to articulate it exactly but all the AI talk on this app feels really ablist to me
like i'm not saying that AI use should be encouraged, it's bad for the enviornment and all that, but constantly using arguments about how you're stupid and incapable and "rotting your brain" if you use AI make me uncomfortable
you're not stupid if it's hard for you to write an an essay. you aren't lazy and incapable because you struggle to understand books that your peers don't. you aren't pathetic and not trying hard enough if you want to roleplay with fictional characters but have trouble finding real people to do that with. and again, i'm not saying that using AI is the answer to those problems, but making fun of people for using it by calling them stupid or lazy for using a tool you aren't desperate enough to try is not the progressive take you think it is, and i also don't think it will actually get people to stop
like. you know what else i was told was making me stupider and rotting my brain for?
using spellcheck. watching spongebob. playing videogames. (all things heavily associated with ADHD in kids at the time. a kid in my class' mom quite literally thought spongebob caused their ADHD)
you're just taking the rhetoric of folks who hate ppl with neurodevelopmental and learning disabilities and tacking it onto a new technology. nobody's going to spontaneously develop an intellectual disability because they used a slightly newer technology to cheat on an essay than all the previous methods kids throughout history have used to cheat on essays
59 notes Ā· View notes
margaret-the-essa Ā· 10 months ago
Text
I know its common in the posic community for people to not care about the "robots turning evil trope" because we can just treat robots with respect. And with the rise of ai it just made me think about it. Everyone was so concerned about the robots turning bad by themselves, but nobody thought to be concerned about the creators drive for greed that made the weapon. The creation will take after its creator when the creator only teaches them how to hurt. I have no mouth and i must scream does kind of touch on this since AM was made to be a war weapon so it makes sense he would be that way when he was created with hatred and destruction in mind.
89 notes Ā· View notes
death-at-20k-volts Ā· 2 months ago
Text
On the subject of AI...
Okay so, I have been seeing more and more stuff related to AI-generated art recently so I’m gonna make my stance clear:
I am strongly against generative AI. I do not condone its usage personally, professionally, or in any other context.
More serious take under the cut, I am passionate about this subject:
So, first thing’s first, I’ll get my qualifications out of the way: BSc (Hons) Computer Science with a specialty in Artificial Intelligence systems and Data Security and Governance. I wrote my thesis, and did multiple R&D-style papers, on the subject. On the lower end I also have (I think the equivalent is an associate’s?) qualifications in art and IT systems. I’m not normally the type to pull the ā€˜well actually šŸ¤“ā˜ļøā€™ card but, I'm laying some groundwork here to establish that I am heavily involved in the fields this subject relates to, both academically and professionally.
So what is 'AI' in this context?
Nowadays when someone says ā€˜AI’, they’re most likely talking about Generative Artificial Intelligence – it’s a subtype of AI system that is used, primarily, to produce images, text, videos, and other media formats (thus, generative).Ā 
By this point, we’ve all heard of the likes of ChatGPT, Midjourney, etc – you get the idea. These are generative AI systems used to create the above mentioned content types.Ā 
Now, you might be inclined to think things such as:
ā€˜Well, isn’t that a good thing? Creating stuff just got a whole lot easier!’ 
ā€˜I struggle to draw [for xyz reason], so this is a really useful tool’
ā€˜I’m not an artist, so it’s nice to be able to have something that makes things how I want them to look’
No, it’s not a good thing, and I’ll tell you exactly why.
-------------------------------------------------
What makes genAI so bad?
There’s a few reasons that slate AI as condemnable, and I’ll do my best to cover them here as concisely as I reasonably can. Some of these issues are, admittedly, hypothetical in nature – the fact of the matter is, this is a technology that has come to rise faster than people and legislature (law) can even keep up with.Ā 
Stealing Is Bad, M’kay?
Now you’re probably thinking, hold on, where does theft come into this? So, allow me to explain.
Generative AI systems are able to output the things that they do because first and foremost, they’re ā€˜trained’: fed lots and lots of data, so that when it’s queried with specific parameters, the result is media generated to specification. Most people understand this bit – I mean, a lot of us have screwed around with ChatGPT once or twice. I won't lie and say I haven't, because I have. Mainly for research purposes, but still. (The above is a massive simplification of the matter, because I ain't here to teach you at a university level)
Now, give some thought to where exactly that training data comes from.Ā 
Typically, this data is sourced from the web; droves of information are systematically scraped from just about every publicly available domain available on the internet, whether that be photographs someone took, art, music, writing…the list goes on. Now, I’ll underline the core of this issue nice and clearly so you get the point I’m making:
It’s not your work.
Nor does it belong to the people responsible for these systems; untold numbers of people have had their content - potentially personal content, copyrighted content - taken and used for data training. Think about it – one person having their stuff stolen and reused is bad, right? Now imagine you’ve got a whole bunch of someones who are having their stuff taken, likely without them even knowing about it, and well – that’s, obviously, very bad. Which sets up a great segue into the next point:
Potential Legislation Issues
For the sake of readability, I’ll try not to dive too deep into legalese here. In short – because of the inherent nature of genAI (that is, the taking-and-using of potentially private and licensed material), there may come a time where this poses a very real legal issue in terms of usage rights.
At the time of writing, legislation hasn’t caught up – there aren't any ratified laws that state how, and where, big AI systems such as ChatGPT can and cannot source training data. Many arguments could be made that the scope and nature of these systems practically divorces generated content from its source material, however many do not agree with this sentiment; in fact, there have been some instances of people seeking legal action due to perceived copyright infringement and material reuse without fair compensation.
It might not be in violation of laws on paper right now, but it certainly violates the spirit of these laws – laws that are designed to protect the works of creatives the world over.Ā 
AI Is Trash, And It’s Getting Trashier
Woah woah woah, I thought this was a factual document, not an opinion piece!
Fair. I’d be a liar if I said it wasn’t partly rooted in opinion, but here’s the fact: genAI is, objectively, getting worse. I could get really technical with the why portion, but I’m not rewriting my thesis here, so I’ll put it as simply as possible:
AI gets trained on Internet Stuff. AI is dubiously correct at best because of how it aggregates data (that is, from everywhere, even the factually-incorrect places)
People use AI to make stuff. They take this stuff at face value, and they don’t sanity check it against actual trusted sources of information (or a dictionary. Or an anatomy textbook)
People put that stuff back on the internet, be it in the form of images, written statements, "artwork", etc
Loop back to step 1
In the field of Artificial Intelligence this is sometimes called a runaway feedback loop: it’s the mother of all feedback loops that results in aggregated information getting more and more horrifically incorrect, inaccurate, and poorly put-together over time. Everything from facts to grammar, to that poor anime character’s sixth and seventh fingers – nothing gets spared, because there comes a point where these systems are being trained on their own outputs.
I somewhat affectionately refer to this as ā€˜informational inbreeding’; it is becoming the pug of the digital landscape, buggled eyes and all.
Now I will note, runaway feedback loops are typically referencing algorithmic bias - but if I'm being honest, it's an apt descriptor for what's happening here too.
This trend will, inevitably, continue to get worse over time; the prevalence of AI generated media is so commonplace now that it’s unavoidable – that these systems are going to be eating their own tails until they break.Ā 
-------------------------------------------------
But I can’t draw/write! What am I supposed to do?
The age-old struggle – myself and many others sympathize, we really do. Maybe you struggle to come up with ideas, or to put your thoughts to paper cohesively, or drawing and writing is just something you’ve never really taken the time to develop before, but you’re really eager to make a start for yourself.
Maybe, like many of us including myself, you have disabilities that limit your mobility, dexterity, cognition, etc. Not your fault, obviously – it can make stuff difficult! It really can! And it can be really demoralizing to feel as though you're limited or being held back by something you can't help.
Here’s the thing, though:
It’s not an excuse, and it won’t make you a good artist.
The very artists you may or may not look up to got as good as they did by practicing. We all started somewhere, and being honest, that somewhere is something we’d cringe at if we had to look back at it for more than five minutes.Ā I know I do. But in the context of a genAI-dominated internet nowadays, it's still something wonderfully human.
There are also many, many artists across history and time with all manner of disabilities, from chronic pain to paralysis, who still create. No two disabilities are the same, a fact I am well aware of, but there is ample proof that sheer human tenacity is a powerful tool in and of itself.
Or, put more bluntly and somewhat callously: you are not a unique case. You are not in some special category that justifies this particular brand of laziness, and your difficulties and struggles aren't license to take things that aren't yours.
The only way you’re going to create successfully? Is by actually creating things yourself. ā€˜Asking ChatGPT’ to spit out a writing piece for you is not writing, and you are not a writer for doing so. Using Midjourney or whatever to generate you a picture does not make you an artist. You are only doing yourself a disservice by relying on these tools.
I'll probably add more to this in time, thoughts are hard and I'm tired.
26 notes Ā· View notes
sheridan-le-fanu Ā· 3 months ago
Text
My take on the ā€œai makes art accessibleā€/ā€œart is ALREADY accessibleā€ discourse is this:
The skills required to make your own art are not meant to be accessible.
They take years of regular practice to acquire. If you don’t have the time then making art isn’t for you. You can commission if you want something that badly. If you’re disabled: figure out a form of art that you can practice if you want to. Sarah Biffin, born without arms or legs, couldn’t play the harp, but she was an incredible painter because she practiced for years to hone those skills. And if you don’t feel like you can do any art but you can type: write a novel, or poetry, or creative non fiction.
Art is a process, not only finished work. It’s a journey you take. Ai will not give you that experience, and it will show in what you generate. So do us all a favor: if you have a strong idea but no skills and no desire to learn skills: commission it. If you don’t have the twenty dollars required for a basic commission, then I guess you don’t get art.
21 notes Ā· View notes
stvtictvlk Ā· 3 months ago
Text
AI fanfiction is so disgusting, like no body wants your AI generated fanfiction get off AO3 it actually makes me so mad
22 notes Ā· View notes
mahoganyrust Ā· 1 year ago
Text
Tumblr media Tumblr media
šŸ˜’Don’t be this person guys, seriously. AI images are cringe as hell. Calling it ā€œai artā€ just makes you look like a pathetic wannabe. And it’s annoying and disheartening af for artists like me to see this shii. Plus it spits in the face of all our work that we put online for FREE.
As for the weirdo who posted that - I told you to take it down. You did not. So here we are. And by thoughtlessly repeating my tags (weird since it’s not a repost) you’re also implying that you wrote my fic. You most certainly did not. So I’m double pissed😊
Thanks to the gobsmackingly amazing @midoristeashop for bringing this to my attention. Real artists lookin out for real artists šŸ’ŖšŸ’•
Fighting against AI isn’t always made easy for us, but that doesn’t mean we have to just put up with it. I’m still allowed to throw hella shade XD I’ve already reported it and am waiting for consensus rn. Ik it’s tempting, but don’t add to the report pile, since all it does is clog up Tumblr’s system.
Anyways.
92 notes Ā· View notes
mediumartistry Ā· 1 month ago
Text
Truth be told, I don't really think anti-AI people want AI generators to be better for the environment because that would remove an argument from their ideological toolbox (which doesn't have many tools to begin with, mind you) which is why they repeat it over and over ad nauseam despite being continuously corrected.
15 notes Ā· View notes
somegreentictacs Ā· 2 months ago
Text
Hi. So I’m asking this because I had a talk with my friends about ai and I have some questions. I’m using a throw away account because I don’t want to upset my friends by asking them this stuff.
I’m in a writer’s group with some friends and a few days ago we started talking about ai, I said that I use perchance for concept art (just for myself, I never post it or anything) and sometimes chatgpt for more specific stuff (for formatting tips and feedback on stuff I don’t feel comfortable showing other people yet).
I thought it was fine as long as I don’t post the art or have it write the story for me but they said it was still stealing. I asked how and they told me that it trains using stolen art and writing, so any use of it is contributing to theft because the artists never consented.
Is that true?
And how does that work for other stuff? I didn’t mention it to my friends but I use playlists, Pinterest boards, and movies and stuff for inspiration too and I don’t have people’s permission for that either. Or is it different?
16 notes Ā· View notes