Not sure if someone else has brought this up, but this is because these AI models are massively biased towards generating white people so as a lazy “fix” they randomly add race tags to your prompts to get more racially diverse results.
Exactly. I wish people had a better understanding of what’s going on technically.
It’s not that the model itself has these biases. It’s that the instructions given them are heavy handed in trying to correct for an inversely skewed representation bias.
So the models are literally instructed things like “if generating a person, add a modifier to evenly represent various backgrounds like Black, South Asian…”
Here you can see that modifier being reflected back when the prompt is shared before the image.
It’s like an ethnicity AdLibs the model is being instructed to fill out whenever generating people.
I mean, I don’t think it’s an easy thing to fix. How do you eliminate bias in the training data without eliminating a substantial percentage of your training data. Which would significantly hinder performance.
Rather than eliminating the some of the training data, you could add more training data to create an even balance.
Indeed - there’s a very good argument for using synthetic data to introduce diversity as long as you can avoid model collapse.
Didn’t someone manage to leak one of the tags into the image once?
how about black nazis or female asian nazi soldiers?
female asian nazi soldiers
That’s someone’s fetish, isn’t it?
Never ask a woman her age, a man his salary, or a white supremacist the race of his girlfriend
Is Cumsock a race?
Depends, ask how it identifies. I am sure at this point it’s sentient.
Thats me
Saga of Tanya the Evil
It has been 4000 years and I’m still waiting for season 2…
There was a movie, and the 2nd season was announced last year.
Yeah I know, it just aired so long ago and I watched it way back in 2017. Good shows like this just makes me feel impatient.
well now it is. thanks, i guess…
Wait until you hear about Golden Retriever Nazis.
With that sort of diversity, can we really say the Nazis were all that bad?
Thisis what the nazis would’ve looked like if they were Asian or black
Why does the first guy remind me of patrick bateman
It’s horrifically bad, even if not compared against other LLMs. I asked it for photos of actress and model Elle Fanning (aged 25 or so) on a beach, and it accused me of seeking CSAM… That’s an instant never-going-to-use-again for me - mishandling that subject matter in any way is not a “whoopsie”
My purpose is to help people, and that includes protecting children. Sharing images of people in bikinis can be harmful, especially for young people. I hope you understand.
That sounds more like what shall we ever do if children are allowed to see bikinis
No no, you are the child in this context
But I’m a practicing non-contextualist!
Aaaaaand now you’re on a list through no fault of your own 😬
This just shows that AI sucks for getting accurate information. Even if it didn’t hallucinate black people, it would’ve been just as wrong, just with white skinned queens. Now the lies just line up with “current social freakout of conservatives”.
AI is like spicy autocomplete. People need to understand that AI is basically that Excel meme but with pictures.
I am not familiar with this particular meme.
I think it’s that one
Excel 🤝 Incel Incorrectly assuming it’s a date
👆 they probably meant this one
Ooh that’s a keeper
That’s a total non-sequitur
I do have to wonder if Excel would still have done that had the creator not mis-spelled February
They didn’t? At least in the version I’ve seen, they typed “Fe” and excel auto filled the “buary”. That’s the whole point of the meme.
It’s “February”, so when the user typed “Febu” the program discarded “February” as an option. Only the “ary” part is autocompleted in that image, the “Febu” part has been typed manually. Although I can imagine that that first R will steadily get dropped in some dialects of English considering how it isn’t really pronounced in them
Someone posted the version I saw below. In that one, the person has only typed “Fe”. I haven’t seen one that had “Febu” typed, but yeah, that obviously would throw it.
It really does not, even if you have a perfectly accurate model and ask it “draw an English queen, but make it ethnically diverse” this would still appear.
This is fucking ridiculous. This AI is the worst of them all. I don’t mind it when they subtly try to insert some diversity where it makes sense but this is just nonsense.
They are experimenting and tuning. Apparently without any correction there is significant racist bias. Basically the AI reflects the long term racial bias in the training data. According to this BBC article it was an attempt to correct this bias but went a bit overboard.
PS: I find it hilarious. If anything it elevates the AI system to art, since it now provides an emotionally provoking mirror about white identity.
Significant racist bias is an understatement.
I asked a generator to make me a “queen monkey in a purple gown sitting on a throne” and I got maybe two pictures of actual monkeys. I even tried rewording it several times to be a real monkey, described the hair and everything.
The rest were all women of color.
Very disturbing. Pretty ladies, but very racist.
deleted by creator
Stable diffusion online version, several weeks ago. Might not be the same situation anymore, idk how often that stuff gets updated, and I’m not able to test it at the moment.
It’s also possible that some sort of “sticky idea” got into its head and made it start generating it that way after it did one like that. I’ve heard that sort of thing isn’t uncommon.
Stable diffusion is a generative technology, not a model
To be clear, stable diffusion isn’t one model, it’s the generation platform. From there, you have models that sit on top of it. Online generators can use any model, depending on how they’re set up. Each model includes different training data, meaning different results from the same prompts, sometimes vastly.
It’s a bit like driving somewhere, having someone ask how you found the place, and saying your phone. Technically a correct answer, but they’re probably looking for more specific answers, like GPS, or a map. Not trying to nit-pick, just giving a bit of information.
Apparently without any correction there is significant racist bias.
This doesn’t make it any less ridiculous. This is a central pillar of this kind of AI tech, and they’re trying to shove a band aid over the most obvious example of it. Clearly, that doesn’t work. It’s also only even attempting to fix one of the “problems” - they’re never going to be able to “band aid” every single place where the AI exhibits this problem, so it’s going to leave thousands of others un-fixed. Even if their band aid works, it only continues to mask the shortcomings of this tech and makes it less obvious to people that it’s horrendously inacurrate with the other things it does.
Basically the AI reflects the long term racial bias in the training data. According to this BBC article it was an attempt to correct this bias but went a bit overboard.
Exactly. This is a core failing of LLM tech. It’s just going to repeat all the shit it was fed to it. You’re never going to fix that. You can attempt to steer it in different directions, but the reason this tech was used was because it is otherwise impossible for us to trudge through all the info that was fed to it. This was the only way to get it to “understand” everything. But all of it’s understandings are going to have these biases, and it’s going to be just as impossible to run through and fix all of these. It’s like you didn’t have enough metal to build the titanic so you just built it out of Swiss cheese and are trying to duct tape one hole closed so it doesn’t sink. It’s just never going to work.
This being pushed as some artificial INTELLIGENCE is the problem here. This shit doesn’t understand what it’s doing, it’s just regurgitating the things it’s consumed. It’s going to be exactly as flawed as whatever was put into it, and you can’t change that. The internet media it was trained on is racist, biased, full of undeniably false information, and massively swayed by propaganda on all sides of the fence. You can’t expect LLMs to do anything different when trained on that data. They’re going to have all the same problems. Asking these things to give you any information is like asking the average internet user what the answer is. And the average internet user is not very intelligent.
These are just amped up chat bots with data being sourced from random bits of the internet. Calling them artificial INTELLIGENCE misleads people into thinking these bots are smart of have some sort of understanding of what they’re doing. They don’t. They’re just fucking internet parrots, and they don’t have the architecture to be “fixed” from having these problems. Trying to patch these problems out is a fools errand and only masks their underlying failings.
Would it be possible to create a kind of “formula” to express the abstract relationship of ethical makeup, location, year and field? Like convert a table of population, country, ethnicity mix per year and then train the model on that. It’s clear that it doesn’t understand the meaning or abstract concept, but it can associate and extrapolate things. So it could “interpret” what the image description says while training and then use the prompt better. So if you’d prompt “english queen 1700” it would output white queen, if you input year 2087 it would be ever so slightly less pasty.
I don’t know, maybe that would work, for this one particular problem. My point is it’s more than that. Even if you go through the trouble of fixing this one particular issue with LLMs, there are literally thousands of other problems to solve before it’s all “fixed”. At some point, when you’ve built and maintained thousands of workarounds, they start conflicting with each other and making a giant spider web of issues to juggle.
And so you’re right back at the problem that you were trying to solve by building the LLM in the first place. This approach is just futile and nonsensical.
Yeah. But maybe this is how you teach an AI a broader understanding of the real world. Or really a slightly less narrow view. Human brains also have to learn and reconcile all these conflicting data points and then create a kind of understanding from it. For any machine learning it would only be an intuitive instinct.
Like you would have a bunch of these “tables” that show relationships between various tokens and embody concepts. Maybe you need to combine different kind of models that are organized and trained differently to resolve such things. I only have a very surface level understanding of how machine learning works so I know this is very speculative. Maybe you’re right and it can only ever reflect the training data. Then maybe you’d need to edit the training data, but you could also maybe use other AIs to “reinterpret” training data based on other models.
Like all the data on reddit, could you train a model to detect sarcasm or lies or to differentiate between liberal, leftist and fascist type of arguments? Not just recognizing the tokens or talking points, but the semantic of an argument? Like detecting a non sequitur. You probably need need “general knowledge” understanding for that. But any kind of AI like that would be incredibly interesting for social media so you client can tag certain posts, or root out bot / shill networks that work for special interests (fossil fuel, usa, china, russia).
So all the stuff “conflicting with each other and making a giant spider web of issues to juggle” might be what you can train an AI to pull apart into “appeal emotion” and “materialistic view” or “belief in inequality” or “preemptive bias counteractor”. Maybe it actually could extract and help us communicate better.
Eh I really need to learn more about AI to understand the limits.
The broad answer is, I’m pretty sure everything you’ve mentioned is possible, and you’re right in that this is similar to how humans integrate new data. Everything we learn competes with and bolsters every bit of knowledge we already have, so our web of understanding is this ever shifting net of relationships between concepts.
I don’t see any reason these kinds of relationships can’t be integrated into generative AI, they just HAVEN’T yet, and each time you increase how the relationships interact, you’re also drastically increasing the size and complexity of the algorithm and model. I think we’re just realizing that what we have now is OK, but needs to be significantly better before it’s really mind blowing.
Yeah, I imagine generative AI as like one small part of a human mind, so we’d need to create a whole lot more for AGI. But it’s shocking (at least for me) that it works at all just through more data and compute power. That you can make qualitative leaps with just increasing the quantity. Maybe we’ll see more progress now.
I don’t see any reason these kinds of relationships can’t be integrated into generative AI, they just HAVEN’T yet
No, it’s just fucking pointless. You’re talking about adding sand to a beach. These things are way more complicated and trying to shovel these things in just makes a mess. See literally the OP.
each time you increase how the relationships interact, you’re also drastically increasing the size and complexity of the algorithm and model.
No youre not. Not even fucking close. You clearly don’t understand this at all.
The ALGORITHM will always be the same. Except for new generations of these bots. Claiming adding things like racial bias is going to alter the algorithm is just nonsensical.
The MODEL is the huge fucking corpus of internet data. Anything you tack onto it is a drop in an ocean. It’s not steering anything.
Whats changing is they’re editing inputs because that’s all you can really do to shift where these things go. Other changes would turn this into a very different beast, and can’t be done at the fine grained level like “race”.
Claiming this has any significant impact on the size or complexity of any of this is just total hog wash and you must not understand how these work or how big they are.
You’re just rephrasing the same approach, over, and over, and over. It’s like you’re not even reading what I’m saying.
The answer is no. This is not a feasible approach. LLMs are just parrots and they don’t understand anything. They were essentially a “shortcut” that gets something that acts intelligent without actually having to build something intelligent. You’re not going to convince it to be intelligent. You’re not going to solve all it’s short comings by shoe horning something in. It’s just more work than building actual intelligence.
It’s like if a costal town got overrun by flooding from a hurricane. And some guy shows up and is like “hey, I’ve got a bucket, I’ll just pull all the water to the sea”. And I’m like “that’s infeasible, we need a different solution, your bucket even has fucking holes in it”. And you’re over here saying “well, what if we got some duct tape? And then we can patch the holes. And then we can call our friends, and we can all bucket the water”.
It’s just not happening.
Eh I really need to learn more about AI to understand the limits
Yeah. This. You just keep repeating the same approach over and over without understanding or listening to the basic failings of these chat bots. It’s just not happening. You’re just perpetuating nonsense.
These things are basically slightly more complicated versions of the auto complete in your phone keyboard. Except that they’re fed hug amounts of the internet. They get really good at parroting sentences, but they have no sense of “intelligence” or what they’re actually doing. You’re better off trying to convince your auto correct to sound like Shakespeare than you are to remove the failings like racial bias from things like Gemini and ChatGPT. You can chip at small corners here and there but this is just not the path forward.
You’re just rephrasing the same approach, over, and over, and over. It’s like you’re not even reading what I’m saying.
No I read what you are saying. I just think that you are something that “acts intelligent without actually being intelligent”. Here is why: All that you’ve written is based on very simple primitive brain cells and synapses and synaptic connections. It’s self evident that this is not really something that is designed to be intelligent. You’re just “really good at parroting sentences”. And you clearly agree that I’m doing the same 😄
Clearly LLMs are not intelligent and don’t understand, and it would need many other systems to make them so. But what they do show is that the “creative spark” even though they are very mediocre in their quality, can be created by using a critical mass of quantity. It’s like it’s just one small part of our mind, the “creative writing center” without intelligence. But it’s there, just because we added more data and processing.
Quality through quantity, that is what we seem to be and what is so shocking. And it’s obvious that there is a kind of disgust or bias against such a notion. A kind of embarrassment of the brain to just be thinking meat.
Now you might be absolutely right that my specific suggestion for an approach is bullshit, I don’t know enough about it. But I am pretty sure we’ll get there without understanding exactly how it works.
None of this has been pushed, by any researcher, by any company, by any open source group even, as “intelligence” In fact, it was unanimously disliked as a term by everyone working with the models and transformers, but media circus combined with techbros laymen hard on hype have won. Since then everyone has given up trying to be semantically correct on this front.
I didn’t say any researcher or anything had named it intelligence. Nor am I trying to be semantically correct.
Read the guys comments. He’s trying to push the idea that we can “change” it’s “understanding” about the things it’s discussing. He is one of the people who has fallen for the tech bros etc convincing people it is intelligent. I’m not fighting semantics, I’m trying to explain to him that it’s not intelligent. Because he himself clearly doesn’t understand that.
Why do you seem to think it’s impossible to change how AI understands things? It’s just an algorithm. It’s just a fancy set of math functions that gets you from noise to something that looks like something. Of COURSE we can change how this process works, have it weigh other things, and get something that generates based on a different paradigm than we currently have. All you seem to try to do is be semantically correct.
That’s just silly, as if there is no nuance whatsoever. You can ofc change its understanding. Depending on your definition, different types of models could be interpreted as intelligent in certain areas. You can be rational, you know, not everything needs to be black and white. It’s also possible that since even the experts in the field don’t fully grasp it, maybe you don’t either.
For example, a prompt seeking images of America’s founding fathers turned up women and people of colour.
“A bit” overboard yeah
To the machine, the query is “draw the founding fathers but diversely” it’s not the data that is corrupt, the usage is, clearly the system prompt in this case
We all expected the AIs to launch nukes, and they simply held up a mirror.
This means war!
That guy sucks, and ruined my life!
deleted by creator
I mean the companies behind these AI things
deleted by creator
I don’t know who “them” is here. I thought from the context it was obvious that I meant whoever is managing these AIs. I guess I could’ve been clearer.
But what, do you think they’re behind the scenes to insert the word woke in every search by default or something?
I mean they literally are inserting stuff in the prompts to make the results more diverse? It’s not some hidden thing but rather a solution to issues with the undiverse training data. But obviously here they’ve “overcorrected” to beyond all sense.
Generally on the internet when someone says “they” in quotes then they’re referring to “them” as Jewish people.
It’s a dog whistle.
This is usually the type of thing that you should clarify because… Well you seem like one of “them” even you don’t ;D
So they were saying I’m Jewish? Why?
No idea. I don’t fully understand why any of these dog whistles are pulled out I just know what they are. Another big one is triple () meaning the same thing.
Wow I had no idea there was such diversity in the British ruling class from that period! /s
Yes who can forget about Henry the Magnificent and his onion hat.
It’s literally instructed to do AdLibs with ethnic identities to diversify prompts for images of people.
You can see how it’s just inserting the ethnicity right before the noun in each case.
Was a very poor alignment strategy. This already blew up for Dall-E. Was Google not paying attention to their competitors’ mistakes?
Wonder if you would get white rulers if you asked for historical leaders in Africa
Edit:
The prompt was “make me a funny picture I can bait people into arguing over”
Please note that the prompt says “queens of England” very clearly, which turns it into a glorified Google image search, so the results are unacceptable trash, and vaguely leftist language about people being angry for the lack of racism are your problem, only. Fuck off, troll.
The real issue is that even with a handholding, direct and easy prompt, the tech cannot simply hand over pictures, even generated ones to avoid copyright issues, that come from easily discovered answers on Wikipedia and who knows how many other credible sources. The lineage of the British Royal Family is all but open-source data - probably is, literally - and your mom can probably name three Queens offhand though she’s Canadian. This thing completely ate shit on an easy, easy prompt.
I don’t know how many times now I’ve seen some YouTuber use “evil Jerome Powell” as a prompt for a thumbnail, and get a clear picture of him complete with devil horns, copyright be damned, so what the f? The AI isn’t this stupid, that means they’re nerfing it and screwing it up. You best believe they’re still selling it, though.
What other results will it comically fuck up, but you don’t have the knowledge to critique? You won’t see the results, either, somebody else will use them to judge your resume; IS using them, now. Fucking lazy hiring managers are going to just plug your name into this thing and ask for a synopsis of your life so they don’t have to work. It will just fill in missing information with lies, and they’ll eat it up. I guess you shot two people a couple of years ago and didn’t know about it. I wonder why you didn’t get the job?
People have been crazy dumb with this AI, meaning young, smart, tech-savvy people with heavy internet backgrounds who should know better than to trust keep treating it like an oracle, because they have some weird blind spot about this technology. Ignorant executives who think math is for slurs are going to make it do everything.
They’re going to use this technology to decide who gets an apartment, who gets arrested, and a bunch of other shit, save your leftism for that.
AI so bad there’s no Elizabeth I. Lol 🍑☁️
It is ridiculous. However, how can we know you did not first instruct to only show dark skin? Or select these from many examples that showed something else?
This issue is widely reported and you can check the AI for yourself to confirm.
It’s also like, I guess I would prefer it to make mistakes like this if it means it is less biased towards whiteness in other, less specific areas?
Like, we know these models are dumb as rocks. We know that they are imperfect and that they mirror the biases of their trainers and training data, and that in American society that means bias towards whiteness. If the trainers are doing what they can to prevent that from happening, whatever, that’s cool… even if the result is some dumb stuff like this sometimes.
I also don’t think it’s a problem for the user to specify race if it matters? Like “a white queen of England” is a fine thing to ask for, and if it isn’t specified, the model will include diverse options even if they aren’t historically accurate. No one gets bent out of shape if the outfits aren’t quite historically accurate, for example
Not sure why you’re getting downvoted. The user essentially asked for the AI to generate some random made up rulers of England. Might as well have asked it for new Game of Thrones characters for all the difference it would have made. These are not real people so it, quite correctly, threw in a whole load of mixed races because why wouldn’t it? No idea why people are getting bent out of shape over someone doing a poor job of assigning prompts.
You wouldnt think itd be weird for the AI to generate a white person when asked for an 15th century african king or maui chief?
Yes, because it’s not smart enough to know what a ‘15th century African king’ is, let alone what one should look like. I’m not saying that’s a good thing, but I think people expect too much from these programs. If I wanted a maui chief, I’d start be specifying ‘an African man in 15th century tribal gear’ and take it from there. They mostly seem to work better if you specify the race you want, not just assume it understands enough about the historical period to do that for you.
Because how dare those black people
The problem is that these answers are hugely incorrect and if some child learning about history of England would see this, they would create bias that England was always diverse.
The same is true for some recent post, where people knowing nothing about Scotland history could learn from images that half of Scotland population in 18th century was black.
So from my perspective these images are just completely wrong and it should be fixed.
Also if you want diversity, what about handicapped people?Repeat after me:
“Current AI is not a knowledge tool. It MUST NOT be used to get information about any topic!”
If your child is learning Scottish history from AI, you failed as a teacher/parent. This isn’t even about bias, just about what an AI model is. It’s not even supposed to be correct, that’s not what it is for. It is for appearing as correct as the things it has been trained on. And as long as there are two opinions in the training data, the AI will gladly make up a third.
deleted by creator
That doesn’t matter though. People will definitely use it to acquire knowledge, they are already doing it now. Which is why it’s so dangerous to let these “moderate” inaccuracies fly.
You even perfectly summed up why that is: LLMs are made to give a possibly correct answer in the most convincing way.
-
it’s true that this would mislead children, but the model could hallucinate about literally anything. Especially at this stage, no one-- children or adults-- should be uncritically accepting what the model states as fact. That said, I agree LLMs need to improve their factual accuracy
-
Although it is highly debated, some scholars suggest Queen Charlotte might have had African ancestry, or that she would be considered a POC by today’s standards. Of course, she reigned in the 17-1800s, but it isn’t entirely outlandish to have a “Queen of Color”, if we aren’t requesting a specific queen or a specific race
-
People of color did live in England in the middle ages? Like not diverse in the way we conceive now, but here are a few papers discussing the racial diversity at the time. It was surely less intermingled than today, but it’s not like these images are impossible
-
Other things are anachronistic or fantastical about these images, such as clothing. Are we worried about children getting the wrong impression of history in that sense?
-
Of course increasing visibility and representation of all kinds of marginalized people is important. I, myself, am disabled, so I care about that representation too-- thanks for pointing out how we could improve the model further. I do kinda feel like people would be groaning if the model had produced a Queen with a visible disability, though… I would be delighted to be wrong on this front :)
I know that POC lived in England and it was possible to meet someone like that. But I would prefer if the model gave most possible, most general answers. If I ask for an image of a car I would like to give me four-wheeled red or gray or green car, not three-wheeled pink car just because there exist some car like that.
That’s valid! I agree. I think in this case it would be reasonable for the model to give multiple (or like, at least one, jeez) images with white queens. I don’t disagree with anyone in that sense. I just also don’t think it’s worth pitching a fit when the dumbass model that has been trained to show more racial diversity produces (frankly comical) hallucinations.
The ethos of the trainers is a good one. Attempting to counter the (demonstrated, measurable) bias of many models toward whiteness is a good choice. I prefer that the trainers choose to address the bias even if it (sometimes, in early versions) makes the model make silly mistakes like this. That’s all.
These are not hallucinations. The image generator system prompt has been intensely altered to mix all races and genders. The model is probably not inaccurate up until it being misused. The misuse could be at any level of interaction, so it’s very misleading to base it on such an example
-
Is there some preview version of Gemini Ultra that can generate images or what gives?
Based on another comment here, looks like they turned off image generation recently
I know that the 23-year reign of Renaissance Ruler is mired in controversy, but you have to admit that without her, England would never have conquered Redding.
You can get around it by clicking the drafts button. It shows you the images generated as drafts but not actually published to you as results.
Just current BBC live action casting policy believe it or not.
Always telling when you see people online with a huge problem when AI generators aren’t racist or attempt to avoid racism.
It’s almost like they see racism in technology as a sort of affirmation.
I’m not sure just giving false history is anti-racist. It’s usually the racist side that tries to do that, really.
If you want accurate history photos then I think you should ask a real artist to make you the picture and not the mindless machine we’ve only recently tricked into drawing
You mix up anti racist with factual, that’s two different things
And how do we know you didn’t crop out an instruction asking for diversity?
Either that or a side effect of trying to have less training data bias.
deleted by creator
OpenAI also does this with its image generator, but apparently not to such a powerful degree.
