Hacker News Reader

Comment by kibibu | Hacker News Reader

Genuine question: what positive use cases are sufficient to accept the harm from image generators?

One that i can think of:

- replacing photography of people who may be unable to consent or for whom it may be traumatic to revisit photographs and suitable models may not be available, e.g. dementia patients, babies, examples of medical conditions.

Most other vaguely positive use cases boil down to "look what image generators can do", with very little "here's how image generators are necessary for society.

On the flip side, there are hundreds of ways that these tools cause genuine harm, not just to individuals but to entire systems.

View on HN

How else do you expect me to illustrate my LLM-generated blog posts about AI?

Oh my. You still make those? Ever since model chupacobra 2.46 we have AI agents making those for us. At one point I was on the fence about totally outsourcing it to agents but it's way more efficient. Now I have 50 posts a day under different names.

Democratizing visual communication is arguably useful, for instance helping people to create diagrams that illustrate a concept they wish to convey. This is contingent on the tech working sufficiently well that the visuals are more effective at communication than the text that went into producing them though.

Can these people not just create a diagram with their own hands? Literally a pencil and paper.

I am at the point where I would prefer a poorly human drawn diagram with terrible handwriting over AI slop.

It is not the making of the diagram that is the problem, but often the fact I have no idea how to put it visually. AI is awesome at this.

Now, does that justify the harm? Not for me, but this issue is way out of my league.

The point of a diagram is that you have something in your head to turn into the diagram. There's no point if you can't do it yourself and the image generator is coming up with it for you.

I disagree. Diagrams are a type of visual communication, and not everyone is good at translating things to visual. I open an excalidraw with clear concepts in my head, but nothing comes out of it. I try C4 or flow diagrams, and I spend an excessive amount of time refactoring them to end up mediocre anyway. Not just me, I know MANY developers that are amazing at explaining things but are mind-blocked when drawing simple circles and arrows.

Helping us navigate things we aren't good at has been one of the main selling points of AI.

Learn how to draw simple circles and arrows, this is the epitome of learned helplessness.

But then he wouldn't have a justification for AI companies to rob people! And you are suggesting robbing himself of this justification!

It's not translation if it's completely AI generated to begin with. Instead of addressing your mental deficits (which sound severe), you're offloading it and making the problem worse.

You have lack of practice and OCD if you’re constantly refactoring but “always end up mediocre”.

I do that. My slide decks these days are hand scribbled.

If you scroll far enough down the linked page, you’ll see they’re knocking off poor handwriting too!

It's always felt like way overhyping to call something "democratization" when it's something I could do as a middle schooler in 2005. It takes some skill to do very well but it's not like basic diagram creation isn't something people already could do for basically free (I create figures for my job all the time now and chatGPT is more expensive than tools I use for design).

Commissioning high quality diagrams from a designer is expensive and I guess it's much cheaper now to essentially commission something but idk, "democratization" still feels weird for just undercutting humans on price.

Yeah, it's not "democratization", people were just too lazy to do it before. It only takes some basic effort and a little bit of time to be able to create decent versions of those things.

You are making a mistake a lot of people make when talking about genAI helping others do work. I get that to you it is very easy to do, but there are other groups of people that are not able to do it. What you are saying is like a hobbyist carpenter saying that making a bedside table would take him one weekend to do, so he doesn't think it is okay for tables to be made via assembly line instead of hiring a carpenter to do it.

I think you're missing my point, which is pretty narrow here. "Democratization" is fairly grand term implying that the general public now have access to something freeing they didn't before (it generally invokes some idea of liberation, as the term often is used to note a transition from an authoritarian to a democratic government). I don't think there has ever been a particularly high barrier to making good diagrams, in my experience it's an easy to learn skill both in time and money, so it feels like it's cheapening the term "democratization". Maybe I'm being a bit sensitive though because of how the world is right now with people sometimes literally fighting for democracy. Normally I am pretty lax with semantics but I've had some people really rub me the wrong way when overhyping AI.

We’re taking about diagrams, all it takes is an experimentation and an iota of thought.

And yet coming up with insightful diagrams, even or especially if they are particularly simple, can be a point of fame (c.f. Feynman diagrams). Diagrams often need to "lie" in some sense, so it can actually be quite difficult to find ways to convey the point you want without misleading in some other important way. e.g. I had a geometry professor that would label the x-axis R^n and the y-axis R^m for a bunch of different pictures, which on its face makes no sense, but it conveyed what it needed to.

People tried to prove the parallel postulate redundant for thousands of years because they lacked the right picture to show why it's necessary.

We’re talking about people using LLMs because they can’t do it by hand. They won’t be splitting atoms with or without LLMs.

My workplace does this for EVERYTHING. And they are always immediately obviously AI slop, both because we all know they wouldn't ever pay an actual artist to create graphics, but also because the people creating the graphics have no sense of style and let it generate the most generic shit possible with zero creativity.

It's definitely not helpful. It's just annoying and disgusting and a waste of resources IMO. But hey at least Powerpoint presentations have AI slop instead of stuff taken from Google Images!?

I'm not convinced that "arguably useful" is sufficient to offset its much more heavily-used application as a casually-available disinformation engine.

I mean, the cat's out of the bag; but the cat stinks.

Democratizing visual communication? Kim Jong Goon has taken away your pen and paper?

The same question could be poised of art in general. I know that response would (and probably should) ruffle peoples' figurative feathers, but I think it's worth considering. A lot of art isn't "necessary for society".

The question still stands, "are the benefits worth the cost to society", but it bears remembering we do a lot of things for fun which aren't "necessary for society".

I used to think like what you describe, but I've fallen on the side of "art is just more emotionally resonant human communication". And most of the time human communication with more effort and thought behind it. AI art falls short on both being human and, on average, having more effort or thought behind it than your general interaction at the supermarket.

I will say, it can be emotionally resonant though - but it's a borrowed property from the perception of human communication and effort that made the art the models were trained on.

The difference between "art in general" and this is scale and speed. Sure, I'll grant you that people are going to engage in deception with or without this but the barrier to entry with this is literally on the floor. Do you have a $5 prepaid VISA? You can generate whatever narrative you want in 30 seconds. Replace the $5 Prepaid VISA with the pocketbook of a three letter agency and it starts getting crazy.

>starts getting crazy

Got pretty wild w/the Iranian propaganda that reportedly _resonated with Americans_ (didn't verify that claim)

Slopaganda - https://www.newyorker.com/culture/infinite-scroll/the-team-b...

Art is for the producer, and if they feel it’s necessary for them to produce it than it’s necessary for them, and what is necessary for the individual extends to the society they’re in.

If you want to say the complete destruction of truth is worth it because some people are having "fun" then idk.

I was worried about the complete destruction of truth, but it seems that's not the result of commoditized image generation. False AI-generated images have been widespread for years, and as far as I've seen, society has adapted very well to the understanding that images can't prove anything without detailed provenance. I'd argue that this has been helped, actually, by random people on the Internet routinely generating plausible images of events that obviously didn't happen.

>society has adapted very well to the understanding that images can't prove anything without detailed provenance

Donald Trump is the president of the United States.

I don't understand the response. Do you think that Donald Trump would not be president of the United States if powerful image models hadn't been invented? Or perhaps you're referring to the AI-generated media he's often posted since being elected; when he showed a video of getting in a fighter jet to dump poo on protesters, do you think many people believed that was a real thing he actually did?

I'm more reacting to the premise that society is positively adapting to the post truth world. Which it clearly is not. Half the population of the US is already living in a fake news mirror universe where everything is inverted. More convincing fake news is not going to help.

And this is just straight out of Putin's playbook, if everything is fake then people just stop beliving in the concept of truth altogether.

I think it's neither going to help nor hurt. My experience is that today, even people "living in a fake news mirror universe" understand that an image does not prove anything unless you can explain where you got it from and why anyone should believe it's authentic.

You shouldn't have believed photos since Stalin had Yezhov airbrushed out of them. The only thing that makes a photo more trustworthy than a painting is that it "looks" more real, and passes itself off as true. But there have always been photographic fakes, manipulation and curation of the photos to push a message. AI will finally end this and people will realise that the image of the thing is not the thing itself.

You are vastly, vastly underselling what is being lost. You can no longer look at a piece of art without first asking "is this even real", that is a collosal loss to the experience of being human. You can't just appreciate anything anymore without questioning it.

>You shouldn't have believed photos since Stalin had Yezhov airbrushed out of them.

It isn't just about propaganda photos, it is about -litearlly everything-, even things people have no incentive to fake, like cat videos, or someone doing a backflip or a video of a sunset.

I agree, but if you enjoy the art, why does it really matter who made it, like I enjoy looking at sea shells, no one made them, but they are nice to look at?

The problem is I'd prefer access to near-photorealistic image gen to be commodified vs something that is restricted, as then only those willing to skirt the law or can leverage criminal networks have access to it.

Every technological advance in this space has caused harm to someone.

The advent of digital systems harmed artists with developed manual artistic skills.

The availability of cheap paper harmed paper mills hand-crafting paper.

The creation of paper harmed papyrus craftsmen.

The invention of papyrus really probably pissed off those who scraped the hair off thin leather to create vellum.

My point is that in line with Jevon's paradox there is always a wave of destruction that occurs with technological transformation, but we almost always end up with more jobs created by the technology in the middle and long term.

Ok, but the models only know what to draw because we fed them images of dementia patients and babies.

Maybe image generators can be a loophole for consent legally, but it seems even grosser morally.

I have plenty for you:

- package design

- pictures for manuals and guides

- navigation and signs

- booklets, tickets and flyers

- logos of all sorts

- websites

- illustrations for books

And many. many others. Not every image is art and very few illustrators are artists.

How do these justify the costs to society?

The 'costs to society' are massively overblown, and some of them (automating jobs) are actually benefits to society.

The girls that have to deal with their classmates generating nudes of them for the rest of time are glad to hear that their concerns are "overblown".

Nobody tell those girls about Photoshop, or scissors and glue.

Nothing says benefiting society like increasing unemployment, destroying what little trust was left in society, and allowing for CSAM and racist propaganda to be generated en masse. At least some corporations will save a few bucks.

Society is fine, maybe its just you who aren't.

So the benefits are that something that was already being mass produced with no issue is slightly easier to mass produce?

It's not a particularly compelling argument.

No, the benefits are that something can be mass produced magnitudes faster and easier, which in turn also creates more latitude for creativity and new spaces.

It's a true state-change, which makes the argument pretty compelling IMO.

Weird how it's the least creative people who use it then.

No idea why you were down voted, I think that's exactly how this will get used.

I'm already imagining this is how the local live indie band night I sometimes go to will generate poster images each week for the bands that are playing, whether to put up at the venue or post to social media. And the bands might be using it to design images to put on their t-shirts and other merch. I already know some indie bands using this stuff for their album covers.

Downvotes because nobody actually wants this. Those image uses serve a purpose to an external audience. The audience doesn't want this shit.

Now of course I'm being dramatically absolute. I'm sure I already consume these things without knowing it. These things serve a function. Offloading to AI is the implementer admitting they can't be bothered to care whether it serves the function.

> Downvotes because nobody actually wants this. > The audience doesn't want this shit.

Speak for yourself.

He's getting downvoted because none of these supposed "benefits" outweigh the costs.

sounds like a very efficient way for an indie band to lose all of their street cred at once

Prototyping. Suppose you have a hard time expressing your vision in words or executing it visually.

1. Generate 100s or 1000s of low-fidelity candidates, find something that matches your vision, iterate.

2. Hand that generated image off to a human and say, "This is what I'm thinking of, now how do we make it real?"

Important: do not skip the last step.

You audit thousands of genAI prototype candidates?

Is the argument any different replacing the word "image generators" with "photoshop" ?

Scale matters. Using Photoshop took vastly more time and skill to pull off realistic images, limiting how many could be made. With image generation there's no practical limit. Some of it will be used for relatively innocuous purposes like making joke images for friends or menus for restaurants. But the floodgates are open for more socially negative uses.

If you're the only one in the world with an internal combustion engine, the environmental impact doesn't matter at all. When they're as common as they are now, we should start thinking about large-scale effects.

It turns out that effort matters

Not much beyond food, water, and shelter is "necessary" for society, but it's nice to have nice things.

I'm teaching my 4 year old to read. She likes PAW Patrol, but we've kind of exhausted the simple readers, and she likes novelty. So yesterday I had an LLM create a simple reader at her level with her favorite characters, and then turned each text block into a coloring page for her. We printed it off, she and her younger sister colored it, and we stapled it into her own book.

I could come up with 10 3 word sentences myself of course, but I'm not really able to draw well enough to make a coloring book out of it (in fact she's nearly as good as me), and it also helps me think about a grander idea to turn this into something a little more powerful that can track progress (e.g. which phonemes or sight words are mastered and which to introduce/focus on) and automatically generate things in a more principled way, add my kids into the stories with illustrations that look like them, etc.

Models will obviously become the foundation of personalized education in the future, and in that context, of course pictures (and video) will be necessary!

Repetition rather than novelty is good for learning.

Sure, and she gets that, but at some point she completely memorizes the stories. She also asks if we can get new books at the store, but they don't make 'em that fast.

Isn’t that also a valuable life lesson that some topics/resources are scarce and at some point you need to do something else?

Sure, and she already got that lesson when there literally weren't more. Then she got another lesson: we can just make our own. In fact that may be one of the most important lessons to learn: you have agency, and you can use the tools you have as accelerants to better yourself, further increasing your agency.

So the use case is just IP theft so you can get more Paw Patrol?

AI aside, if you’ve truly exhausted all the simple readers, maybe she should move on to more advanced books instead of repeating more of the same and gamifying it, which seems a great way to destroy a child’s natural curiosity.

Sure, I don't view "IP" as valid, don't entertain the idea that it is possible to "steal" it, and absolutely don't care that someone out there might be sad imagining me making a coloring book for my kids. In fact I'd go so far as to say that holding the position that there's something wrong with tailoring teaching to a child's interests and avoiding that for fear of copyright concerns of all things actually makes you morally bad.

You overestimate how many there are. There's like 10 stories at that level. I do also read ones with paragraphs to her, but she can't do those herself because she's 4.

Ah the old sovereign citizen reverse uno. It's actually evil NOT to use the art theft machine to dumb down your children.

Yes, generating tailored practice material for a continuous difficulty curve and to keep their focus with something they enjoy is dumbing down. Exactly.

Do you get this upset at illegal drug users for their flouting the law (e.g. recreational marijuana is still illegal everywhere in the US) as you do with me making reading material for my own children? Do you get this upset at artists themselves who no doubt "stole" others' art (e.g. copied a drawing or drew a character they did not "own") at some point in their learning process?

I also sing Raffi songs to my children without asking for permission! I hope he doesn't mind!

That is not IP theft, that's private use. If (s)he tries to sell those coloring books, that's then theft. You're free to do anything you want with IP in privacy, it's only when selling or exhibiting to the public IP law is triggered. Knock yourself out with protected IP in private.

You're thinking of fair use, and that's the worst interpretation of it I've seen.

But it's true. You can do anything you want with private IP in private. It is the dissemination and distribution of IP that not yours that is the issue.

Not even remotely true. Fair use doesn't give you licence to pirate, unless you're a politically connected AI company.

It is not piracy to acquire private IP legally (someone has to get it in the first place) and then you can to anything you want with it in your own privacy. It becomes an issue when your activities with that private IP is no longer private. think it through, I really don't think you have. BTW, I'm CTO of a law firm.

CTO of a (immigration) law firm lol. Bro you’re not a IP lawyer

Okay, so I have looked this up, again. I was an expert in this area for work purposes around the 2000-'10. Checking the law now, the definition of fair use in the US has changed, been modified to extend "IP stored in digital files" as protected even in private. However, it is pointed out that there is no means of detecting such private uses, and any actions enabling the detection of this private use are no longer private and fall right into the previous IP law.

Saving money for businesses trying to promote their products?

> Genuine question: what positive use cases are sufficient to accept the harm from image generators?

Diagrams and maps. So much text-based communication begs for a diagram or a map.

Could the same argument not be applied to practically everything and have drastically different perspectives from people?

people pay them to use it, they find that positive

There are many use-cases outside of spam and slop.

For example, take a picture of your garden. Ask chatgpt to give you ideas how to improve it and a step by visual guide.

Anything that can be expressed visually is effectively target for this technology - this covers pretty much everything.

Are those sufficiently valuable that the death of photographic evidence is worth it?

That's a multimodal model with text output, I think GP is asking about image generators.

I a 5’5” male can make myself look taller on dating apps

Short kings on tinder no more!

Reply on news.ycombinator.com