AI-Generated Images Discourage Me from Reading Your Blog | Hacker News Reader

AI-Generated Images Discourage Me from Reading Your Blog

nelson.cloud

796 pointsmeysamazad465 comments

To me there are different motivations to include imagery and some of them are just fine with AI generation. If the imagery actually goes to the substance of helping to illustrate or explain the actual content then it seems very superficial to reject it based on how it was generated. In fact it is kind of stupid to insist someone slavishly draw something by hand that could have been done an easier way.

The problem comes when the imagery is placed as a false value signal. If it is designed to send a signal of "this content took a lot of effort to produce because an image that would be hard for a human to make is attached to it" - but the image was generated without that effort, then it's basically a lie and insult to the reader.

This. If it's actually an illustrative part of the text, include it, even if it's AI generated. Not everyone who can write can do graphics.

If it's only decoration, leave it out. No matter whether AI, stock or custom made.

not everyone who can think and solve interesting problems can write prose well, but there's still a lot of complaining about AI-generated text.

One of the reasons is that, most of the time, LLMs can't do graphics or write prose well enough either; otherwise it wouldn't reek of LLMs so much to people who put some effort to read it. Also, even poor writing or drawing can be elevated with some intentionality and personality, and using LLMs pretty much guarantees blandness.

You can use AI as part of your editing process. But it needs to be a process, back and forth with human input, and human editing at the end. What people primarily complain about it pure AI text. It has sentences a paragraph long using every punctuation mark invented by man. If you look at book you'll notice the same style of punctuation used but it'll be at 10% of the rate. You won't see n/m-dashes, commas, colons, semi-colons, and parentheses all used in a single sentence.

One of the most tedious ideas of recent years is the "great engineers are great writers" meme, unsurprisingly spread by writers.

Unfamiliar with this meme.

But I'd still rather read a badly human engineer written blog post than an LLM generated one.

Me, too. Plus, who cares if an LLM corrects the grammar or something.

If the engineer's a bad writer, the blog will at least restrict itself to the elemental information. There's too much SEO-influenced fluff anyway with people telling their supposed life stories as introduction to some technical tidbit. Same pestilence nowadays as with cooking recipes.

Nobody wants to read that shit, no matter who or what generated it.

For most people your blog is really for you, and if anyone is reading it on the regular than bonus points... but the vast majority of the planet who runs a blog the traffic is very low.

So do with it as you wish, if you enjoy playing with AI and having it assist you with any part of it it's your corner of the net do with it as you absolutely wish and everyone else can kick rocks.

Most artists would prefer you to steal their art than paying a multi billion dollar company to steal it for you and make a worse inaccurate copy. And there are petabytes large free or cheap archives of clip-art, diagrams, and illustrations all over the internet.

For every AI snob that's not reading an article because they see an AI image, there's 10 non-snobs who appreciate a decent usage.

If you're creating complete slop, no one is reading your article in the first place. Full stop. The image doesn't matter.

As with anything, it's about what you created, and creating at all - not the tool you used to create.

The snobs and haters will continue to b!tch and moan. They always have. Always will. Your two cent elitist opinions are still only worth two cents.

> it's about what you created

But you didn't create anything in this case. Not the images at least. Which then casts doubt on the rest of the writing in said blog post.

There is an artistry in assembling other art into a single cohesive piece. The film director doesn’t design the sets or give the performances or do the makeup, but they place an authorial stamp on the production by selecting what will and will not end up in the frame.

> There is an artistry in assembling other art into a single cohesive piece

I never said there wasn't, but I also wasn't talking about human roles where people in larger creative teams have some kind of artistic say in the final output of a much larger process involving other talented humans. That is a very different example and I think you know that.

Oh yeah I can just remember all those famous renowned collages that everyone associates as art.

A director does a hell of a lot more than select what goes in a frame and then sit back and watch it come together.

A collage made by Picasso is not associated as art?

There is artistry in assembling other art into a single cohesive piece. A film is a collection of the work of artists and performers. Combining AI imagery and AI text together does not make art, because none of the constituent parts are art. A few prompts do not make art.

Writing a blog post is a creative effort, selecting a few pieces (generated or not) is creative, too, whether it is up to your standards are not.

The gatekeeping on this is crazy. I really wonder how artists discussed photography and other advances.

If you write a blog post, yes that is creative. If an AI does it, it's no longer creative, it's derivative (by definition). Selecting AI images for one's blog is technically creativity (akin to decorating), but in the most hollow way possible.

If you compare this scenario to photography and art, it's pretty clear you've not actually participated in that community

Somebody still has to prompt it, there is a creative task. Somebody still has to curate it, that too is a creative task.

I was comparing this to when photography first came onto the scene not today, but I'm tired of the gatekeeping (which you retreated back into) which says this isn't art when as with every creative process, it is derivative and creative.

People already splice together videos of generated clips. Is that not creative? It's just gatekeeping. People in a few years (or already) or in a generation won't be this pedantic about it.

If you use the definition of "creative" in that someone creates, sure, you can make your argument. Most people, however, include a level of imagination and effort in their qualifications of whether something is creative. An AI prompt does not meet the bar of "creativity" for most people. Curating AI slop similarly does not meet this bar. Your personal bar may well be on the floor, but that is not representative of most people, in my experience.

You're using your definition of "creative" and "art" interchangeably, which lacks nuance and respect for art. The reference to the arrival of photography similarly lacks nuance and respect for artists and art.

Art and creativity have a human element, they evoke emotions. As soon as you delegate the process to an LLM, you've disrespected that human element if you try to pass it off as creative or art. You can call it "gatekeeping" if you want to, but I see it as basic respect for other humans and human imagination and effort.

Yeah I just don't think there is anything special about "art" and I don't think there is some specific definition for what art is. People create that's art. You can put different amounts of time and effort into it but I think it's just as lacking in imagination or respect for human creativity to distill this down to he worst case.

If somebody spends hours prompting various resources and curating it, is that then "slop"? I don't think it is, it's just as sustained a creative effort, regardless of the tool used. You are saying that the 10-20 minute clips that people have put together of generated vid of various series (like Star Wars) isn't art? There are entire stories told using this method. It's not art?

To summarize, this is a "you" problem. Just because YOU do not think there's anything special about art, it does not mean there isn't. Your lack of respect for art does not undermine the thing itself. You cannot have your GenAI products included with "art" simply because you want to. You aren't entitled to having your prompt outputs held in the same esteem.

> If somebody spends hours prompting various resources and curating it, is that then "slop"?

In my opinion, yes. Some people may disagree with the term "slop" in this case, but that does not step it all the way up to "art". It can still be non-slop, non-art.

I mean sounds like it's a your problem if you are saying that certain tools, no matter how they are used or the effort put into using them can't qualify as art.

You're right, I don't gatekeep art that same way. I don't lack respect for art, I lack respect for your standards which I think are artificial.

GenAI only exists because it steals from a vast corpus of human creative output. It then regurgitates it back, stripping away the emotion, care, and talent that made it art. No amount of prompting will touch the amount of accumulated skill that the original work required. If you cannot understand why trying to pass off GenAI output as "art" is disrespectful and illogical, I cannot help you.

Someone that pretends prompting is "effort" will always just be that, regardless of how you, personally, set your standards.

> Your lack of respect for art does not undermine the thing itself.

Don't you see how this can be said for your feelings about AI art?

Everything you're saying about AI art, someone once said about some other emergent art form. Photography, CGI, digital illustration, synthetic music... with just as much conviction as you're displaying here. But you'd agree those other things have become artforms, yes?

To preempt the typical retort "Yeah but in those cases humans put in some extra X" okay why can't humans put in some extra X while using AI as a tool as well?

What other people have said about other, unrelated things has no bearing on what I'm saying about this particular thing. The illogical conclusions of nameless "others" are not an argument against my current conclusions. You're treating me as guilty by association.

You're also making an assumption that AI's acceptance WILL follow, just because the others were later accepted. You can't make that assumption at all.

Neither of these arguments are in good faith.

How are they not in good faith?

They are logical fallacy and deliberately do not engage in the actual argument. It's an inauthentic engagement.

Right now your actual argument is just really your opinion. You can avoid actually examining it closely by claiming logical fallacies, but again we're just talking about your opinion, you're not making a logical argument.

The colloquial definition of "creative" and "art" are not solely an opinion, it's a tangible thing you can explore and investigate if you're motivated to do so. The inability of LLMs to express creativity is also not an opinion. The differences between film directors and LLM prompters are also real things that we could discuss.

But a meaningful discussion requires good faith, not dismissal to disingenuous argument.

You're acting as though the ordinary meaning of the words settle the controversy, but that's exactly what's contested here.

Whatever definition you bring into scope will not resolve it because a colloquial definition only describes how a word is commonly used, and doesn't establish rule for every disputed case, which is what we have here.

Definitions are generally not normative, but even if we agreed on one, it would not exclude a works solely because of the tool used to produce it. But you are saying that every work produced using AI is illegitimate as art, regardless of its intention, composition, context, expressive content, or any other quality ordinarily associated with art.

What necessary property does accepted art possess that no AI-assisted work could possibly possess? Or to state it another way closer to my original question (which was not disengenuous), what additional human contribution makes photography or CGI art, and why couldn’t an AI-assisted process contain that contribution?

Until that property is identified, saying AI images cannot be art is largely just a stated conclusion.

The definition of "creative" and "art" are, indeed, contested in this discussion. Establishing a definition is, and has always been, a large part of many debates. We don't have to necessarily agree, but at least be able to state our position and discuss from there.

Instead, you did not engage with this concept until now and relied on logical fallacy. When I pointed this out, you did not revise or adapt to show good faith, but decided to pivot to falsely classifying the discussion as just a matter of my opinion with no merit. Now we're finally circling back.

There are a number of necessary qualities missing in AI works to qualify as "art". Let's take photography, as an example. In photography, the camera captures the images for a human, but the "art" is the unique (original), intentional depiction of the scene that captures the artist's perspective. Art conveys the artist's thoughts to others. This is true of CGI as well. It is still the artist using their skills to provide a unique perspective in a digitized world.

AI works lack these qualities. It is impossible to create unique AI art - it isz at best, a collection of derivative parts. Sometimes art is intentionally derivative, yes, but that leads to the next point. AI art cannot carry intentionality that art can. It carries a distillation of intention, tokenizing and associating the intention back to its training set. It disconnects the prompt author from the output in a definitionally important way. It is randomized, not reproducible. It is an approximation of the prompter's thoughts, ideas, and emotions - not a translation. It cannot reflect the human experience in its output, just a statistical guess at them.

> saying AI images cannot be art is largely just a stated conclusion.

Yes, it's called a thesis. I have one, I've stated it, and I've supported it. You're welcome to contribute to the discussion

I pressed all the letters with my meat fingers to express my human thoughts and it's still not genuine enough. shrug

I don’t think you understand the meaning of the word gatekeeping.

Please, elaborate.

Is a painting not art because individual pigments aren't art?

I'm not sure if you're being sarcastic or not... I hope so.

Assuming you aren't, this is called a "false analogy" or "category error". Pigments are raw materials for painting in the same way that letters are raw materials for writing. Pigments make no claim to be art, never have. AI-generated text, on the other hand, is a fully-realized product that lacks the human intent that makes it art, but can be passed off as "art".

I think they would obviously be happy with the alternative wording "it's about what was created" or just "it's about the product". I don't think they care about any aspect of how it came to exist.

Getting an AI to create the imagine you want isn't necessarily easy.

But you didn't create anything if you used Photoshop... You didn't create anything if you used synth and autotune... You didn't do any scientific research if you used a calculator. You didn't do anything if you just piped together two programs.

I've heard it all before.

Gate keepers wanna keep the gate as long as they can.

Don't forget photography. Photography is going to destroy art, don't you know.

You probably put a fair amount more effort and intention in than from just writing a text prompt, I'd wager. All of those examples would require you to at least have some kind of idea about your intended output.

> at least have some kind of idea about your intended output.

Isn’t that what a prompt literally is?

Why do you presume people who use AI don't have some kind of idea about the intended output?

These statements are factually untrue or overly broad to the extent of being nonsensical. Hopefully, you can see how that would come off as incredibly disingenuous. [E.g. what did they autotune? What did they Photoshop? Probably some creative work that they or somebody else made.]

[Edit: Further, these don't really line up with what the comment you're responding to is saying. If all you did was write a prompt for an image, it gives the impression that you may have done the same for the text. At that point, if I care about what the AI has to say, I can just write a prompt to an AI myself.]

> What did they Photoshop? Probably some creative work that they or somebody else made.... If all you did was write a prompt for an image

Most serious AI artist are not just writing a prompt for an image. They usually have their own process and can describe it in the same detail any trad artist would.

But to your point about Photoshop/autotune, the basis of most AI artwork pipelines is some other creative work, like a reference image, just like you're imaging is the basis for those other examples. So the argument isn't disingenuous, you're just not familiar with how AI artists work.

For example, if an AI artist wants an exact composition for a scene, they can first create that composition by taking a photograph, or arranging it in a 3D editor like blender. Then they can use AI to "fill in" some elements of it.

I don't see that as any different from an artist using autotune to make their voice sound different.

Moreover, what about the synth example? What goes into synth music is connecting a bunch of wires and adjusting a bunch of knobs to tune the generated output of the machine. That's the same mechanical process as AI art.

I use AI all the time and do some paid work as an artist. Although I don't use AI directly in my own work, I am very familiar with the workflows in 2D, 3D, and video.

> you didn't create anything in this case. Not the images at least. Which then casts doubt on the rest of the writing in said blog post.

The topic is "AI-generated images" in blog posts. Which, no, most of those aren't being generated with any of the more sophisticated techniques you're listing. And if they are, they're less likely to be the images people are complaining about and that are being discussed here.

People are complaining that search engines are overwhelmed by AI-generated writing and images. Several people in this thread are stating that they click-off when they see it. It may be developing similar to banner ad blindness, where people saw the pattern of spam and low quality output and developed a reflexively mental filter and stronger dislike of them. What number of potential readers is an acceptable loss? Most blogs are small, so probably not many. But even worse, as many have said here, it also harms their perception of the blog.

Outside of the many outright spam blogs generating tons of images, a lot of people creating the "AI-looking" art that people say looks "bad" or "obvious" are actually also putting time and energy into it too. Those things don't automatically equal quality or trust. In some cases they consider themselves the serious AI artists. These creators don't always know that their images "look like AI," which is part of why it's becoming a perceived quality signal. They'll say it doesn't, but people mistake some of my pieces (not AI) as AI. Art is in the eye of the beholder. (Incidentally, if you're reading this, ensure you're actually getting meaningful, critical critique and feedback. That's something that's an important cultural component in many creative communities.)

I take issue with how poorly phrased the comment's arguments are. Saying that people "didn't create anything if they used Photoshop", they didn't create with whatever "synth" is supposed to be, or that scientific research isn't real because of calculators are bad comparisons and inaccurate.

The author may be trying to make a point similar to the photography one mentioned elsewhere or something like "Well if AI generated art isn't art, art must not exist?" I'm literally confused by the post and honestly can't make out the point with how hyperbolic and outright wrong it is. It just reads as mud-slinging to me.

> What about the synth example

Does "synth" mean synthesized music? What about it? A synthesizer is generally deterministic and the musician is making a decision about exactly what nodes are being played. The use of something like an arpeggiator is predictable and deliberate choice (I'm not going to go into the artistic integrity tier list of every synthesizer module.) Text prompt-based generation is a much higher level of abstraction which often makes several compositional and stylistic decisions for the user, even with advanced prompting.

Comparisons like these are unconvincing, which is part of why the other comment isn't good.

For some group of users, images they perceive as "AI" now carry a spam association. So, is your music actually rising above the cacophony? Do people even see your banner ad? Do they instantly get a bad taste in their mouth?

Go ahead and use AI-images in your blog post, I'm not gatekeeping or going to stop you. If they're so great, use a bunch of them everywhere. But, I (and people in this thread or that I know) will assume it's blogspam and click away. And, if it's commercial setting and I'm your competitor, I'm going to view that as potential opening for differentiation.

> The topic is "AI-generated images" in blog posts. Which, no, most of those aren't being generated with any of the more sophisticated techniques you're listing. And if they are, they're less likely to be the images people are complaining about and that are being discussed here.

The person you originally replied to scoped their comment to "decent usage", which to me implied some more process than just prompting.

> Does "synth" mean synthesized music?

Yes.

> A synthesizer is generally deterministic and the musician is making a decision about exactly what nodes are being played.

Generally but not always. There's plenty of synthetic music that leverage stochastic components. The stochasticity is a useful design constraint and is part of the artform. There's nothing that says art has to be deterministic or that the artist can't leverage nondeterminism in their artwork.

> Text prompt-based generation is a much higher level of abstraction which often makes several compositional and stylistic decisions for the user, even with advanced prompting.

There's also a whole genre of synthetic music called "Live Coding" which is typing programs to create synthetic music in front of an audience. There's also nothing that says art is only about making low level decisions.

> Go ahead and use AI-images in your blog post, I'm not gatekeeping or going to stop you. If they're so great, use a bunch of them everywhere. But, I (and people in this thread or that I know) will assume it's blogspam and click away.

It all depends on who you're looking to attract. The people I've met in AI focused communities have been awesome, always looking to experiment, try new things. The people against them I find very conservative and resistant to change (not saying this is you, or that it's necessarily bad). Which is fine if that's your audience, you have to know your readership. It's sort of like the difference between startup culture and enterprise culture. One is known for being nimble and innovative, the other for being reliable and capable. Positive signals to one are anti signals to the other. Threats like "I and people I know and the people in this thread won't frequent your blog" are met with shrugs, because that just means you're not the target audience.

I’ve heard this gatekeeping argument before, but it’s not convincing. Nobody prevents anyone from doing anything with any tool of their choosing; nobody prevents anyone from writing (or doing art or any other business, really). There is no gatekeeping or democratization of art needed either. If you just prompt an image, it has zero value, and if your blog is just a copy-paste from any LLM, what’s even the point?

Prompting an entire article is a waste of time for everyone. Just outsource all thinking already. Oftentimes a generated illustration signals that. Laziness, lack of discernment. Low consciousness, cheating. (Not always of course, sometimes you just want an illustration for your blog, no big deal! Use AI!) That said, writing doesn’t need to be streamlined like a production line. People immediately feel when you haven’t put the work.

This point against “gatekeepers” is a very lazy argument.

And the same can be said about taste. Taste comes from learning and learning is practice. No wonder people are put off by generated “content”.

What if you go through a couple of hours prompting and thinking to get to right image? What is the cutoff to "lazy" or "zero value"?

What if you played Pachinko for two hours? Even if you won two dollars, you basically wasted your time.

So regardless of effort and result, AI use for image creation is wasted time for you?

People have heuristics. My heuristics is that AI slop in images -> slop in the text, whether AI or not.

It will be wrong sometimes, but mostly it's not. OP has a point.

“Everyone who disagrees with me is a hater and a snob.”

Hmm, I wonder why you're taking this so personally...

Indeed, including some AI generated things is becoming a costly signal that you're not bound by the neoluddites and the online anti-AI mob / redditor hivemind. I'm starting to like this signal, but it has to be distinguished from the simply being-lazy signal that indicates the content may not be all that interesting of valued by the author. If you have an otherwise high quality site where you aren't afraid to use AI for some elements, that makes me like it more.

It's a start of a barber's pole of fashion. Like when it became desirable to wear torn jeans even if you can afford proper ones. But you can only safely do it if you won't be confused with an actual hobo. It signals not playing by midwit pearl clutcher rules. So same with AI use. Don't play to the pearl clutchers, but don't be a hobo either. Shower but have somewhat torn jeans.

A key part of this is it has to be conspicuously deliberately AI.

> It signals not playing by midwit pearl clutcher rules.

Pre-torn jeans signal being a poser, wanting the torn jeans look without having the lifestyle that results in torn jeans, or even having the agency to tear your own jeans

So it's an apt analogy for AI users I guess. People who want to build creative works without ever putting in any effort to learn how

Modern society rewards taking shortcuts way too much

I mostly agree but here my analogy was reversed compared to yours. In my analogy the nicely dressed church lady is the professional, politically correct, climate conscious, anti AI redditor. The torn clothes punk hobo is the lazy rulebraking AI shoveler who makes the shared spaces dirty, and the smart iconoclast contrarian rebel artist poet ignoring schoolmarm rules is the person who uses AI but also makes otherwise high quality content.

(don't get hung up on the climate thing - I'm not arguing on the object level. A Christian believer can also be annoyed by the Karen-like peal clutcher no-fun-allowed church lady archetype).

I get what you're going for, I just don't think there's anything remotely punk, contrarian, or rebellious about using AI

AI is the establishment

Context-dependent. Online in social media the backlash is becoming the new purity test.

A slop image header is a signal that the text is probably also slop. Why should I waste my time on the off chance that your text is not actually slop?

You can call it superficial, but I’m not reading your AI blog, I’m not eating at your restaurant with AI food pictures, and I’m not listening to your AI music.

I’m quite happy without all the slop.

Come on. We all know what kind of images this post is talking about. No need to play dumb.

Same problem for me for sending emails, outreach, and other behaviors that signal human identity but it turns out to be an AI post. The area with grace for me is how much effort a human spent on composing the message - I wouldn't expect a message a human wrote >50% of and had an AI help grammar check, proof read and edit need an AI disclosure. But otherwise, if the identity is >50% have an AI disclose it.

This is how I'd like for human-on-behalf-of-human outreach to work too, i.e. if my human secretary sends a message under my identity and via my email client, I'd expect it to be under their name not mine.

This points out some long standing issues with lack of accountability and deception by humans through our time. Think: humans have been doing these things:

- human have other humans impersonate them and send messages on their behalf (i.e. via secretary, receptionist, email, a VA, via social media via a social media manager)

- friends or family answering your letters if your a celebrity

Those behaviors, deceptive and lies, are insults to you at the expense of the gain of the liar. We haven't held them to accountability or a standard of respect before AI, and so we are facing the same challenges we did before AI - we just have more humans doing the same thing with AI as the <insert old human role here>. Why have we been unable to hold them accountable to the deception before - lack if transparency, lack of power to hold accountability to the behavior?

It's like styrofoam stucco on the ceiling.

Decoration and ornament have served always to signal extra effort that went into something, signaling that it is considered important and valued by the maker/owner. But people also have incentive to do cheaper mimicry. And then the viewers have incentive to distinguish real from fake to see where true, scarce effort went. These dynamics are crucial for understanding the feelings and emotions around AI. Value is almost never in the thing itself or some inherent quality but in the social signals and relationships. (somewhat less true in very utilitarian engineering machinery or for sustenance level basic necessities).

That is one use.

Images (incl. icons) can also be there to structure text and provide visual clues on what is where.

In a blog, real images signify you spent a lot of time creating the post, it was very important to you, and you likely made sure the details in the text were accurate.

While AI images in a blog signify low effort high volume posting where the text likely was just copied from AI without any care.

The article author should have specified more closely which type of image. Stock image sites had already made generic photos and drawings cheap to include. And indeed they look cheap. The more airbrushed and high quality the cheaper it looks. If a dentist website has cheap beautiful stock nurse photos, it's less valued than real photos taken by a photographer at their location.

Once in a while you get a good one, one that stands out among all the others, and that's a worse outcome because the same image will be used by everybody including your competitors.

At that point why have images at all?

Just go full American Psycho eggshell business card.

The OP doesn't have images.

It's hard to articulate this stuff, and especially hard to explain it to the AI infatuated among us because they don't really want to understand why people dislike the tooling they're obsessed with. AI is good at many things, digesting vast quantities of information, and code being most of them. Everything else to me looks/sounds like shit. AI music sounds dull and overproduced in a way that a lot of middling pop music does. AI images look overtuned and strange in the same way a magazine editor would, if they had no self control with their tools. It just... looks bad. As an aesthetic. If you are not aesthetically minded, you may not care, or you may not be able to tell, and that's fine. But I am and I can and if I open your blog post and see AI crap, I'm leaving. I'm not asking anyone else to, and I don't give a shit if my best friend in the world wants to beat his meat to your AI generated sunset, good for him, I hope he has a great time. But I'm not interested.

Why these super-pro-AI people take this as such a personal slight is strange to me. You didn't care enough to actually make something, what are you offended if I tell you I don't really want to look at it?

The tools are getting better and often that overproduced thing is the toupee effect, where you don't notice the better ones. The generic look can be avoided with more customized prompting. You can make much more natural songs with longer prompts, a bit of quickly self-sung melodies you add as references, even if you can't sing. Similarly for images, sketching something out in paint can make a huge difference or just iterating 2 or 3 times and not accepting the first result.

And the tools are going to get better too.

> especially hard to explain it to the AI infatuated among us because they don't really want to understand why people dislike the tooling they're obsessed with

Your comment is an example of what the parent comment is discussing; the reflexive "you can't really tell and anyway you can do it better to stop people being able to tell and also soon it will be better so no one will be able to tell" is a gallop like the narcissist's prayer.

People are telling you they can tell, and that they do care. The toupee fallacy works because there are good wigs that no one can identify unless told, but the idea that that there's actually brilliant AI prose (that no one has detected) always lacks examples.

"AI Slop" seems to be misunderstood despite its simplicity: "this is sloppy/low-effort/unskilled work ... enabled by generative AI tools".

> "People are telling you they can tell, and that they do care."

In the same spirit, one could argue that "AI Slop" is actually constructive feedback to a small degree. It's a short-hand for "this is sloppy/low-effort/unskilled work with obvious utilization of generative AI tools". It's not much different than using the same photography filter for every post on Instagram or picking the most popular Wordpress theme --- if you are expecting praise for that then you will be disappointed.

I do deeply enjoy that apparently among gen-Alpha, it's a common thing to say when someone says something really stupid, or makes something of poor quality, to call it "AI."

The kids are alright.

It's fine to care about the human element and human connection. Analogy: It's fine to want to hold a large vinyl in your hand, look at the artwork, caress the sleeve with your fingertips, whiff the smell of the record and enjoy the experience. It's just that just as with "digital CD music is quantized and sounds worse than analog" got exposed as nonsense, the same way "I can tell it's AI and that's why I don't like it" will also not stand the test of time. You can still want human music for the human reason, but you won't be able to tell them apart.

> It's just that just as with "digital CD music is quantized and sounds worse than analog" got exposed as nonsense, the same way "I can tell it's AI and that's why I don't like it" will also not stand the test of time.

I mean you're rather boldly skipping over the part where the former is a proven phenomenon (though I would say it's more nuanced than true or false but there's too much to unpack there) versus something that, as yet, I have not experienced.

> You can still want human music for the human reason, but you won't be able to tell them apart.

I will pepsi challenge any media you like, from the most powerful models you like, happily.

And I'm only willing to even entertain images in terms of said challenge. The music and video are so fucking obvious, just absolutely glaringly AI when they are that I don't feel the need to even dignify it with a test. As yet, I've seen some images that certainly get close, but there are always tells in my experience.

The biggest mistake is thinking that your aesthetics are "correct".

You make it a personal slight when you espouse, that your aesthetic preferences are universal and then farther malign people with different preferences by saying that they are not aesthecticly minded.

Yeah yeah, no aesthetic preferences are ever bad, sure. Heard it, don't believe it. Your thing sucks, sometimes that's just a fact.

> The biggest mistake is thinking that your aesthetics are "correct".

Good thing I didn't say that then.

> You make it a personal slight when you espouse, that your aesthetic preferences are universal and then farther malign people with different preferences by saying that they are not aesthecticly minded.

Aesthetics are just taste and taste is subjective and will vary by person. That said you can, broadly, achieve "good" or "bad" aesthetics based on opinions at scale. The predominant opinion right now is that AI art is bad. You feel what you feel about that, but numerous prominent people, numerous companies who provide services, etc. are all responding to the fact that people find AI art ugly in a variety of ways. Any stock photo site I use has an option to filter AI art, and if it doesn't I don't use it. Projects in the public space that are outed for using it get backlash. Various social media sites are demonetizing generated content.

And to be clear, me saying people are not aesthetically minded is not meant as an insult. Plenty of people I respect do not care how shit looks or feels. But they are a minority in my experience, and frankly, anyone working in an even remotely creative field probably does care. YMMV.

Be that as it may, though, aesthetics do matter to a lot of people, even if they don't discuss it openly or lack the vocabulary to do so. People will say it just "feels wrong," or it's "soulless." It seems to activate our brains very similarly to things deep in the Uncanny Valley, where it looks allllmoooost right, but then fumbles something subtle in a way that makes it off-putting in retrospect.

I don't think we'll ever get rid of AI entirely. I think it will remain in the margins as cheap solution-oriented ways to make graphics, and maybe video again if they can ever figure out how to make it profitable. But I have a feeling an AI video commercial would run on TV in a similar time-slot to recycled Billy Mays ads, and with similar function: cheap, already there, get the job done, and the audience is not discriminating. It's "fine" in the broadest sense of the word.

I don't think it signifies any of those things. This is all new stuff. Some people have decided to offload image-creation to machines because what they care about is the text.

This stuff you say it signifies, it says more about you than them.

If you don't care about the images then don't have images.

Or just have a podcast so that you don't have the option to have images.

It bothers my brain when a longtime followed podcast changes their cover image. You conditioned my brain for years to a certain image and pattern and now you change it? How dare you!

It's so weird that they all seem to update at once - which makes me think that it's my podcast player that's periodically checking. But that sort of an opaque process for me since it's not my area of technology.

No one ever said they didn't want images, this is a non-sequiturs

> Some people have decided to offload image-creation to machines because what they care about is the text.

Thank you for providing the text that reaffirms my statement. If you have the time to copy and past someone else's text, you should take the time to actually read it

It's about experience visiting many sites and spotting a correlation. There is a very bottom of slop, the equivalent of the chumbox from earlier. But on the higher end it may be more like you say. If there are other indicators of quality, I wouldn't penalize it for AI use. It's like having lots of piercings and torn clothes and tattoos. You may be a genius scholar who is kinda into this as a hobby or you may be an actual lowlife who can't hold down a job and gets into trouble with the law all the time.

People will always make snap judgments, you can't inspect every website in detail. So if you use AI, make sure there are some other signals of quality.

Yeah, everyone knows the best content is on a professor’s site from the 90’s, with no images. Maybe some figures, but if so, they should look like they were printed, physically cut out, and then scanned. Bonus points if the background of the image absolutely does not match the color of the webpage.

So now we have this low-effort image generation tool. At least it doesn’t indicate that the author wasted a substantial amount of time looking at pictures. Still a silly waste of time, but at least it is a small one.

Some use of hero images by bloggers is a deliberate style choice, to emulate the norms of professional journalism.

A lot of traditional news outlets have a style rule that every article must have an image. Look at https://www.bbc.co.uk/news https://www.nytimes.com/ https://www.economist.com etc - every article has an image. Even a report on something highly abstract - "BP reports record profits" - will be illustrated with a photograph of some BP-branded fuel pumps. Stories on things like cybersecurity and inflation get generic images like snippets of blurry computer code, banknotes etc.

Some bloggers feel their blog is improved by adopting the style of professional journalism, and (among other things) put an image on every article. Even if for a programming blog good lead images are hard to come by.

Interestingly, professional journalism organisations often have pretty strict codes of practice about images, and photoshop use. They have rules on everything from illustrating one rocket launch with a photo of another to posing the subjects to colour-correcting images to creating multi-image collages. So I don't imagine we'll see many AI-generated images in traditional news outlets any time soon.

> Interestingly, professional journalism organisations often have pretty strict codes of practice about images, and photoshop use.

I seem to recall many instances of the use of incorrect photos, and the earliest I heard of was in the The Guardian in the 1980s before the general decline in media quality and standards.

Yes, the current standards did arise out of a lot of questionable behaviour.

It's awfully tempting for a war photojournalist to pose the group of tearful women in front of the bombed-out building across town. Or for the celeb gossip reporter to collage together two photos without making it clear it's a collage. Or for the person colour-correcting a photo of a divisive figure to lighten or darken their skin.

And the resulting rules are complex and at times judgement-based, because posing subjects is allowed in some circumstances, collages are sometimes allowed, some colour adjustment is allowed, and so on.

Still, serious journalists aren't going to be illustrating their articles about US-Iran negotiations with chatgpt any time soon.

are there any serious journalists any more? If so, who?

I recall some rather awkward/amusing examples of this in media reports on certain video games. It's better now, but at one point, about half the featured images for articles about the game Super Mario World used some random pic from DeviantArt to represent the game (despite said art looking nothing like the original title), while a similar number of articles about Super Mario 64 represented it with a screenshot of the ROM hack Super Mario Star Road.

No-one knows exactly how said images ended up getting this sort of usage. Presumably some random journalist was on Google Images one day, come across pics that looked vaguely like what they remembered the games to look like, and just kinda went with it.

I also remember quite a few cases where fan-made box art mockups got accidentally used as the pictures of the real thing, and when the BBC used a logo from Halo to represent a real-life organisation.

People who know more about the topic than I do will argue for a default image for an online article for SEO and other reasons. Online pubs I've done stuff for would usually take a good/relevant photo if you have one. But, otherwise, would generally use some fairly generic stock graphic from some service they subscribed to.

Same goes for book covers. You probably need something unless you want to look really generic and cheap. But I'm mostly not going to pay a designer hundreds+ of dollars for a cover of an independently published book. I have had a former colleague who is actually good with this sort of thing do it in the past. I've also done a crappy job myself, done it in conjunction with a publisher (for whom I had the supply the cover image), and, yes, used generative AI/Photoshop.

Someone paying hundreds for a good cover design is a signifier that a lot of effort went in to the book and the cover reflects that. An AI cover signifies that this is a quick rush out the door cash grab book that likely isn’t worth reading.

There was a thing about a year ago where YouTubers were holding lapel mics in their hands to show they don't really know what they're doing, on purpose, to differentiate themselves from the big players.

If it looks like cheap generic crap it’s probably cheap generic crap. Most AI imagery looks like cheap generic crap.

It’s flooded the market to the point average users are starting to turn away. Look different or be relegated to obscurity.

It's not just about effort though. I prefer the authors terrible doodles over AI art any day of the week.

Some of my favorite examples:

https://hamberg.no/gtd

https://learnyouahaskell.github.io/introduction.html

It's weird to observe that trend in house decor, first it was mark of wealth, after that when cheaper methods to replicate the look (and some might be even more smooth than doing it by hand) it flew down to cheaper and cheaper had that look, while "wealthy" moved to simpler one, often just showing off amount of space or some expensive material used. And now more simple looks trickled down too

Stucco on the ceiling is done to hide imperfections from hastily installed drywall.

In cheap housing constructing, every step hides the previous steps mistakes. Framing - drywall - trim - texturing - painting - decorating.

A while ago I saw a gif of a piece of art being made. It was pretty ugly and I wasn't impressed.

Then the camera zoomed and I saw that the thing had been painstakingly assembled by hand out of two million toothpicks over the course of years.

Suddenly I was very impressed!

> Value is almost never in the thing itself or some inherent quality.

Value is in the thing itself, however, there requires a certain level of trust by the reader consumer that what they are taking part of is not a scam/bamboozle first, and hopefully worth their time, and some proof of work goes a long way. So effort used to signal that the author cares about the subject if they went through the effort of setting up a blog, writing 2000 words on a topic with some curated CC-licensed images relevant to the subject.

GenAI images and LLMisms are the inverse - proof of no-work; they signal the author didn't put in effort, other than a fraction of a cent on tokens. This is why "cleaning up" your writing with LLMs is not a great idea, or makes it look low-effort.

> These dynamics are crucial for understanding the feelings and emotions around AI

These issues predate AI, going back to content-farms (if not earlier). The root issue is low-quality content masquerading as high-quality and authoritative. GenAI greatly lowers the bar of entry.

>In fact it is kind of stupid to insist someone slavishly draw something by hand that could have been done an easier way.

Communicating low effort used to be insulting to the consumer, yet now I'm expected to praise it? I understand that we're not talking about monetary exchange, but we're still talking about my finite time in this world, time that I could've been invest reading or looking at something else.

So we're going back to my favorite adage these days: If you cannot spend time writing it, I refuse to spend time reading it.

In fact, the moment I realize someone's content is AI generated, I just feed the links (youtube -> transcript, blog posts -> link) to chatGPT and have it generate a summary. That's because if you're in favor of using your time efficiently then so am I.

Should everyone draw their icons by hand, too?

Clipart exists, free use art exists, the Gimp Green Pepper stamp tool exists. I would read a blog with extensive pepper stamp art before something with over the top hacker aesthetic AI generated slop art.

So what about the use of the odd AI generated icon, not some extensive use?

Gotta use the green pepper stamp tool for bullet points. /s

Little icons would be less noticeable at least.

The thing is that if you do it "right" (screen usage, correct when it is needed), it probably won't be noticed as AI in the first place.

> The problem comes when the imagery is placed as a false value signal. If it is designed to send a signal of "this content took a lot of effort to produce because an image that would be hard for a human to make is attached to it" - but the image was generated without that effort, then it's basically a lie and insult to the reader.

Well, the issue is the majority of AI images seems to fall into this category, which then trains my brain to treat all AI images in this way, rather than sifting for the few golden use cases where it's legitimately adding value.

At work, for example, I'm more likely to go brains off if someone presenting content shows an AI generated image in their slides.

The only reason most of my blog posts have an image at all is because social media seems to want some type of thumbnail image to go with the post. If I don't have something, many of them seem to just grab my logo instead which seems...spammier.

I try to find something better when I can, but the need to put something on every post is exhausting.

This is my take. Most of my (https://rich.sh/blog) blog posts, are of actual things I can screenshot or take a photo of, and if I can I'll use that.

But, for abstract posts, random bits of code, notes, etc - it looks bad to have them be all the same generic icon or my headshot for them it seems off-putting. I'm not a graphic artist, so I instead use MidJourney generated icons for social / post icons that match the topic of the post.

No. Even setting ethical considerations aside, AI-generated art is lazy, looks gross, and signals poor taste whilst wasting computing resources. I think it's a perfectly reasonable principle to claim, especially if we're discussing diagrams. I'd much rather look at a low-effort hand or MS Paint drawing, there's more humanness in that than any generative AI turd.

True until LLMs become smarter than humans. At that point, you __want__ the AI generated image because it signals that AI was used.

Best case it won't make a difference and trust is the norm again (imo). Even if artificial super intelligence will be a thing, it will drown in the swamp that is communication just like all super intelligent humans do.

“AI signals” are not a hallmark of quality; it’s the other way around.

A superintelligent LLM/agent that wants its content to be read by humans (…which itself is suspicious - if not ominous) is not going to dress itself up as a slop-farm.

Also I’d like to remind all the AI boosters and LLM glazers in the audience here that if something is interesting enough to read, it will be interesting enough to write; if you outsource production to an agent then it tells me you don’t actually care about the output, and I won’t care to read it.

Not necessarily disagreeing on writing/reading as such - and there is an understanding effect in writing, I think.

That said, do you care in a similar way about important or interesting calculations being done by humans instead of machines? What is the cut-off function machine vs human?

The most important thing to keep in mind, when considering AI images for helping to illustrate or explain the actual content, is that AI-generated images have a big trust problem.

Take a look at the Google Image Search results for 'Engine Diagram' https://www.google.com/search?q=Engine+Diagram&tbm=isch

In the first row I get two correct-looking diagrams, then a load of AI-generated ones. Three cylinder heads for four cylinders? Two crankshafts? Oil pan labelled Crankshaft? Oil pan labelled Connect in rod? A four-cylinder engine with seven valve springs? Parts labelled up to 39 but a 9-item key? Spark plug wires that loop back on themselves? A crank shaft where the pistons aren't offset in phase? Four-cylinder engines whose exhaust manifolds have three or five openings? Then in the second row, I get "Parts of a V6 engine" showing a four-cylinder engine - with an 'Exhaust malvol', an 'extaust manivol', an 'exsoust manivol' an 'oilolaivt manivol'

Savvy technical audiences have noticed this - and so when a diagram has a whiff of AI about it, they become extremely skeptical and distracted.

Personally I only use AI-generated diagrams very sparingly, and very carefully.

To be fair, the problems you describe existed before AI, though I concede it has gotten worse.

What baffles me is why anyone would think it a good idea to post and host these images? What’s that upside?

Same reason fishermen put baits on the hooks, or hunters make weird sounds: it helps attract unsuspecting victims closer, increasing the chance of extracting some value out of them.

> the problems you describe existed before AI

No, the problem of mass-produced, outright wrong, educational graphics flooding image-search results did not exist before 2023.

> To be fair, the problems you describe existed before AI

To an extent, I guess?

But in the past if one wanted a diagram and didn't care about copyright they'd grab an image like https://i0.wp.com/www.engineeringchoice.org/wp-content/uploa... and reuse it unmodified.

In the age of AI, one can hand it to an LLM and get https://mechaniclove.com/wp-content/uploads/2023/12/engine-p... in a different style. And with messed up labels and the wrong number of holes in the gaskets.

> To be fair, the problems you describe existed before AI, though I concede it has gotten worse.

I do think it's fair to say it's gotten a lot worse.

Before AI-generated imagery, an hilariously bad (yet human created) engine diagram as described above, might actually be shared among my friends for laughs. Because it didn't happen that often. Now it happens a lot more often, and they aren't human created (which was only funny for a very short while).

No human drawing an engine would have called an oil pan a crank shaft. A thing that in no way resembles a crank or a shaft.

Diagram-wise I had better luck in telling it to make a text version for diagram generator, not only AI edits tend to be more correct, it's easier to fix any errors too.

As a bonus you can then prompt AI to validate whether the diagram matches article and vice versa, I've caught some small bugs like that when we migrated our docs

> you can then prompt AI to validate whether the diagram matches article and vice versa

No repro with Nano Banana with an infographic prompt.

I used Claude code that I pointed at local Kroki instance for rendering, and made it note how to use it in CLAUDE.md for validating

> In fact it is kind of stupid to insist someone slavishly draw something by hand

I believe they used to call that "art."

No, they used to call that "illustration."

The belief that we were living in a golden age of hand-made visual beauty before AI arrived is obvious nonsense.

The golden age of illustration ended when Illustrator arrived. After that, everything became a mess of cliches and clip art, culminating with Google's Material Design, where everything looks simultaneously soulless, heartless, and bland.

AI brought some colour back. A lot of the imagery is kitsch, but illustration has always tended to kitsch, so nothing really changed there.

This is the irony of the anti-AI purity hunt. Most human-made work is crap. It always has been.

AI hasn't suddenly sent the quality of creativity off a cliff. It's commoditised the low standard of most previous hand-made work and made it easier to access.

The high end - the real talent - still stands out. But most people can't afford it, and even fewer people appreciate it.

>AI hasn't suddenly sent the quality of creativity off a cliff. It's commoditised the low standard of most previous hand-made work and made it easier to access.

Yes. That's why it's called slop. And now we're absolutely swimming in it. I guess a small number of people like it, but then there's no accounting for (bad) taste.

> "The high end - the real talent - still stands out"

It's a shame we made it economically infeasible to ever more produce or pursue.

The value of even mediocre art used to be that it still took a human somewhere hundreds or thousands of hours to reach that skill-level; and its existence indicate true care by the creator or commissioner. Heck, most historic art in cathedrals and historic art museums is crap by modern standards; they simply didn't have the tools and resources to reach high level in the craft.

> "AI hasn't suddenly sent the quality of creativity off a cliff"

No it has. Because there is no point anymore, no economic value, no recognition of the time and effort required. Spend 1000 hours to be called worse than AI. Spend 3000 more to be accused of using AI. Finally spend 100000 more, and finally reach above the slop; only not being paid a dime because AI is 90% good enough. Want to spend those final 300000 to finally get paid for your work? Or do you prefer working minimum wage instead.

Just go to google and use their work as is, and crop out the signature. At-least then you still acknowledge that there actually is a value to creating imagery left. Welcome to the world of slop.

> In fact it is kind of stupid to insist someone slavishly draw something by hand that could have been done an easier way.

Nah, they just forgot to use the "slavishly drawn by hand" prompt!

>Redraw the attached image in the most clumsy, scribbly, and utterly pathetic way possible. Use a white background, and make it look like it was drawn in an old computer painting program with a mouse. It should be vaguely similar but also not really, kind of matching but also off in a confusing, awkward way, with that low-quality pixel-by-pixel feel that really emphasizes how ridiculously bad it is. Actually, you know what, whatever, just draw it however you want.

Source: one of the ChatGPT image presets

> In fact it is kind of stupid to insist someone slavishly draw something by hand that could have been done an easier way.

That is not at all what the author wrote. They said they would rather see a shitty hand drawing than an AI image, they’re not “insisting” you make a hand drawn image. You could just as well use one of the thousands of free stock photos out there just like blogs have been doing for years before AI images.

Or, more likely, they just lifted copyright protected photos from somewhere on the web without even attribution,

No doubt many (I don’t think you can conclusively say “more likely”) did. There’s a crapton of blogposts out there and no one has verified them all. That doesn’t matter one iota for the argument in the submission, though. And it doesn’t really serve as any kind of defence either, when we all know generative AI is trained on a massive corpus of images without permission or compensation, to a much larger scale and much larger impact. Go ask image authors and artists which one they’d prefer: having all their works stolen to feed generative AI or having random bloggers using a single image in a blogpost. At least the latter have a good chance of being attributed to their authors (and if they’re not at the start, they will if the post gets popular enough and gets backlash).

> In fact it is kind of stupid to insist someone slavishly draw something by hand that could have been done an easier way.

Would you also say this about the writing? It seems to me from your description that you consider drawing something to be harder than writing something.

Which is fine, but to be fair, I'd also prefer that people who find drawing easier than writing, to not outsource their writing to an LLM. By which I mean I would prefer not to read that.

> In fact it is kind of stupid to insist someone slavishly draw something by hand that could have been done an easier way.

Its not. Putting in the effort makes sure there was a reason for it in the first place and gives the author to re-evaluate his assumptions.

AI-generated images discourage me from interacting with any entity AT ALL. Bar / restaurant posters, events, shops, anything. Because it makes the brand appear lazy. And in a weird way "anti-human". I don't have a better word than that.

You're setting yourself up for a lonely life.

I think it is like CGI in movies, in the beginning it was often obvious and jarring. Over time as it gets better the obviousness of AI images will be much harder to spot and they will be common without being noteworthy.

> You're setting yourself up for a lonely life.

What a ridiculous conclusion jump.

It's the complete opposite of that. I prefer human things, human interaction, with human beings involved, which in some way reflect the human experience.

> lonely life.

Maybe they are hoping robots will replace human interaction, connection and intimacy like in some of those future distopian movies

I want to say /s but I'm sure lots of people are looking forward to this because they are already lonely

No. This anti-AI attitude is getting old and tired.

I'm getting sick of people insulting the work that I do, calling it lazy (it's not), and saying I should do things in the manner they prefer. I'm also sick of the childish death threats.

If you don't want to interact with me because of AI, good riddance.

> I'm getting sick of people insulting the work that I do

Do you do it, or does AI do it?

If it takes me two days to build a service or make a short film, I'm doing the work. It doesn't will itself into existence.

If I use an excavator to dig a ditch, it doesn't mean I didn't dig the hole.

This has gotta be satire

Somehow we managed to live this whole time without generative AI. Why do we need it now?

"Just eat the shit, everyone is eating the shit, eat the shit with us or you'll be a lonely weirdo"

hm, ok

This is kind of like saying in the 2000s "If you won't listen to auto-tuned music, you're in for a lonely life." There's entire subcultures where you can avoid that, for example, fans of the Grateful Dead (who include many successful people, such as Jerome Powell, but that's another story). Anyway, they don't tend to be lonely people.

Spineless post

Well it just so happens that I am mostly avoiding post 2000 movies and overuse of CGI is a big part of it.

Yesterday I encountered AI-generated food photos on a menu for the first time. I immediately felt grossed out before even realising why. Their generated focaccias didn't even look like food, just off-putting trypophobia triggers.

100%. Uncanny valley but for food. Or anything else.

> Yesterday I encountered AI-generated food photos on a menu for the first time. I immediately felt grossed out before even realising why. Their generated focaccias didn't even look like food, just off-putting trypophobia triggers.

Also, that's just misleading. When I see pictures of food on a menu, I expect them to be real pictures of an example of the actual dish I'd get (perhaps made a little more carefully with more of an eye towards presentation, but the actual, real item).

AI-generated food photos on a menu are dirty lies.

Not that either are OK, but McDonalds basically stages their food to make it look more appealing, adding stuff that isn't in the product to make it look nicer.

https://youtu.be/pSYw8LTXXJM?is=FsJwTFKUkeDt1Z2Y

Tangentially, I kinda like the idea of Doordashs crowd sourced food images. Take a Pic of your food before you eat and get like a $5 discount if it is chosen. Customers are not likely to manipulate the product (they may stage it nicely like you mentioned but won't put browning agent or other additives to make it shinier).

who can forget this scene... https://youtu.be/zJs9p-VNORw?si=sLF7ve6AhrQw8fNn&t=245

McD's is also not considered quality food so I don't think this example goes against the association of fake pics -> no thanks. But the practice also becomes more egregious when applied to more niche foods where the customer doesn't already have an idea of what the food could look like.

> When I see pictures of food on a menu, I expect them to be real pictures of an example of the actual dish I'd get

I don't get who would think that's a good idea. Like if I order something because the picture shows that the dish has carrots or something and the dish you bring doesn't I'm going to complain.

But then again I can't tell you the number of restaurants that have a photograph of a menu on a website where the original was clearly digital already.

>> When I see pictures of food on a menu, I expect them to be real pictures of an example of the actual dish I'd get

> I don't get who would think that's a good idea.

It's super helpful to the customer, because it gives you a much better idea of what you're going to get. That's especially true with ethic food, where the descriptions use unfamiliar terms or a lot of meaning is lost in translation.

> Like if I order something because the picture shows that the dish has carrots or something and the dish you bring doesn't I'm going to complain.

Seriously, what's the problem with that?

Are you, for some strange reason, taking the perspective of a restaurant that wants to behave in a very customer-unfriendly way? Randomly dropping or substituting ingredients is not something a restaurant should be doing.

I think you're misunderstanding me, I'm agreeing with the GP that AI generated food pics are a net-negative.

The sentence is more "I don't get who would think that(AI generated food pics) is a good idea" but I could see how it was ambiguous given the quote.

I’m afraid it will become the norm very soon, and nobody will give a damn about it, even if they notice. Most people already don’t notice it at all.

Last Christmas I caught our municipality (I'm from Greece) posting announcements about Christmas events on brochures that were choke full of obvious AI-generated images with Santa Claus, deer, trees, etc. and my first reaction was wondering exactly how hard would it have been if they were to hire artist(s) from our region, even if it was volunteer work, to have them promote their work instead. That's what a community is about.

> wondering exactly how hard would it have been if they were to hire artist(s)

approximately infinitely more hard, considering using AI is 0 hard.

And hiring and billing are nearly infinitely hard unless you already have a routine set up for it.

It seems a particularly bad idea for bars/restaurants because people are very, very sensitive to food that looks _wrong_. I’m not sure I’ve ever seen an ai-generated food image that didn’t look offputting.

Very much of a survivorship bias. As of today, I don't think you can conclude if the images on UberEats/DoorDash are AI-generated or not. It's not 2023 anymore. Things have gone way past that now. I hate it as well, but I've seen AI-generated photos that I would've never guessed as AI from a quick scroll on a small screen.

Truth be told, they didn't put much effort into it before GenAI.

Those restaurants that use AI menus now? They just used free templates before. Few mom'n pop restaurants could afford spending money on graphic designers to work out a nice menu.

Usually they'd just use a template, send to the printer (or print it themselves), and that's what you got.

No they definitely did not use "templates" for the pictures of the actual food before. They had either actual photographs or no pictures.

There’s also a class of “detailed agent written technical blog” that’s you can just scan and get the shape that it’s long paragraphs, tables of statistics, and dry LLM writing. You notice the blog has one or more such posts a day.

It’s a turnoff immediately if someone shares what their agent wrote.

It also has happened a few times they contain a fair number of misrepresentations and mistakes.

On the other hand, it could be great to have collapsed sections clearly marked as LLM with the name of the model, the prompt, the assumption the author somewhat checked it and didn't use it to write the article.

You can put it inline or at the end of the article: references, citations, further reading, "AI".

(It might also be the only way to stop training models on their own output)

Are you speaking in the abstract, or would you, personally, like that when reading a blog?

Speaking for myself, I have absolutely no interest in reading anything written by an LLM that I didn't prompt myself, under really any circumstances I can think of.

You are thinking of something nice to have. At least half the idea here is that it would be nice not to have something.

Usually there are many buttons and links one wouldn't click on.

If authors can't resist having chunks of copy paste slop in the middle of their article they can do it in a way I don't have to see it.

They also get a fair opportunity to impress me with a prompt that looks interesting enough.

And they get to distance themselves from the idea that copying [say] a Wikipedia article into a blog posting is the same as writing something.

> Are you speaking in the abstract, or would you, personally, like that when reading a blog?

It's better than the current status quo, where people post LLM writing without attributing it to the LLM, hoping you'll get tricked into reading.

At least if it's labeled like the GP describes, you know what you're getting and can make your own decision.

On the other hand:

> (It might also be the only way to stop training models on their own output)

On the other hand, I think it's a moral imperative to make it so the models train on their own output. I have absolutely no interest in giving them clean input data. After reading about how ISBNdb is helping them destroy old pre-LLM books for input, I totally think someone who does not have small children should start creating print-on-demand forgeries of such books (with the same ISBN numbers) filled with the most garbage-quality LLM text possible.

I don't mind it for footnote/reference-book type material, where you want to have some information available for the curious but 99% of your readers won't care.

A few dense paragraphs about a database implementation that's ancillary to a story about the content of the database, for example; a small percentage of readers might want to reimplement or host A Thing locally and it could be helpful to know that it's this type of database with that number of tables and takes up x gigbytes on disk etc.

I don't see a lot of value in that, but one could supply a lot of semi structured extra information which could aid the readers LLM to answer questions the reader might have with more accuracy / factual grounding.

It's a turn off both because you don't know if it's accurate at all, and because LLMs have a way of turning 1 sentence ideas into pages of diluted slop.

Even if it used to be accurate. Percona were a great resource and now it's all giving LLM vibes. Maybe it's still accurate, it probably is, but it also feels like they don't care

I just sat through my first fully AI-generated presentation. The content was as bland and straight down the middle as possible. The slides were all generated by a standard design - a mix of text blocks, bullets, fonts, colors.

The whole thing was packed with information and completely worthless. I’m 100% convinced that before too long we’re all going to be allergic to this low effort crap, and craving actual people with an actual point of view.

this is what I've observed as well, it's supposedly impressive and packed with "information" and .. stuff but it's just kind of noise to me

it's fatiguing to read that stuff as well

Yeah, I have told people that I’m fully capable of typing things into ChatGPT or Claude myself. What I need from people are their own warty ideas. Sometimes real ideas are blurry or incomplete. Or have open questions. Making your own presentation can be a personal exercise in clarifying your thoughts. And sometimes the fact that ideas are half-baked or tricky is exactly why you’re having the meeting in the first place…

I guess if you’re in an environment where the emphasis is put on busy-work and formalities, then you’re going to get this sort of shit. Deeper organizational problem at that point, though…

I used AI to make a presentation that animated around (html). It had the look of the storyboard and the postcards had tweakable controls on them for the data I wanted to show. It was pretty slick, something that would take me forever (and I'd never get the time approval to do).

same here. our org just hired a new guy and he just did his first presentation, blatantly AI-generated. he said everything that had to be said on the first slid, then wasted 20 minutes going in circles.

llm writing style sounds so wrong once you say it out loud.

I’m inclined to agree, but not as much because the image was generated with AI but rather because it’s rare for such images to exhibit taste. They feel like the visual equivalent of gilded twinkies, which is offputting.

Selecting a stock image doesn’t have this effect, even if it’s not related to the post. Even the most unremarkable stock image has the fingerprints of its artist, and the choice itself can reveal things about the author. They don’t have that hyper-polished unintentioniality and subtle uncanny hollow “wrongness” that generated images do.

If an LLM has any taste, it comes from its training data.

LLMs don’t have taste. Almost definitionally they can’t. They can parrot what they think other people think is tasteful, but they cannot have taste.

Taste is about matching things. If you stick Van Gogh's Starry Night into a blog post about profitmaxxing because you know it increases your discoverability by 0.02%, you have no taste. If you stick a pleasingly-proportioned image of a sailboat bobbing peacefully on a calm sea at sunset on a blog post about getting away from it all, it might be a perfect fit.

I haven't done research on this but I feel like a lot of stock image companies are using AI now

Probably correct, which is why date filters come in handy.

I recently complained to the marketing team about an AI generated image used on our website main dashboard. I described everything that was wrong with the image the more you stared at it. That if I was a customer, I would feel that if they are ok to show a low quality image there, what does it means to the products I'm buying.

Their first reaction was to say they didn't see any problem with the image, then they changed it for an abstract image that I couldn't tell if it was AI or not, which is a huge improvement.

I thought I was alone in feeling this way, but turns out people from the sales team told me privately that they were happy I had spoken up.

Public shaming and rejection are probably the best way to teach people it's wasting people's time and disrespectful. If you couldn't otherwise write it yourself for whatever reason it's best not to until you can.

> Public shaming and rejection are probably the best way to teach people

Sigh

> Public shaming...

People already tried to cancel others because they didn't agree with them. Eventually the most of people came to a conclusion that it is no good. I think at least that the most did just that. So I'm proposing another solution, just ignore what you don't like. No one is forcing you to read a blog that uses AI generated images or uses a wrong color for its background.

I don't know what you really mean by "cancelling", but I think it's totally fine (and even good) to talk about the flaws in stuff.

It’s totally fine to talk privately with someone. Public shaming tend to have very negative effects.

> I don't know what you really mean by "cancelling"

It surprises me, that you don't know. But if you wanted to know you'd ask LLM, not me, right? So I'd rather spare myself from explaining it.

> I think it's totally fine (and even good) to talk about the flaws in stuff.

To talk about the flaws and public shame are different thing, don't you think?

> So I'd rather spare myself from explaining it.

Being snarky doesn't really help your position.

> Being snarky doesn't really help your position.

Pity. I laid great hopes on that particular phrase.

To cancel someone was really popular for a couple years starting just after covid. There's probably a better definition somewhere, but basically it is when a collective of like-minded people decide to publicly shame a person until they are too embarrassed to have an online presence, or even worse, until companies are pressured to not employ the person. It was like unofficial unions against someone existing. It stopped after enough people were cancelled that they were no longer powerless together. Then people who hate others had to evolve into different ways to be disrespectful because this method stopped working, which is why you don't see cancelling anymore

I would say that "cancelling" is mostly organized groups of people publicly shaming and harassing third parties to get to their victim. So like demanding that his employer fires him, demanding that any sponsors cancel their contracts, showing up with a mob outside their home to kill their family, demanding that they get removed from platforms like YouTube or Facebook.

This is a bizarrely twisted narrative that does not conform to reality.

>Public shaming and rejection are probably the best way to teach

That's a wild stance. If you don't like what someone else made - AI assisted or not - you can just move on and not purposefully set out to publicly ruin them...

It's good as a community to communicate what we prefer to read. At least that way the author knows why they're getting the cold shoulder.

Why ruin it for the large community who didnt notice it or mind it, and actually like it? Give cold shoulder by literally walking away. Yelling "SLOP SLOP SLOP" is opposite of cold shoulder, no?

I'm fairly AI-pilled, as the cool kids say. But it would be fair for any AI-generated work to be labelled as such. The problem is, vast majority isn't doing it, because they know what the perception is like.

The question is, should we spend the same amount of time evaluating AI-gen work as non-AI-gen work. I would say no. In my opinion, AI-gen work should be evaluated by other AIs. Goes back to the whole "if you didn't put effort into creation, why should I put effort into evaluation" thing. Obviously it's not an all-or-nothing thing. But you get my point.

>Public shaming and rejection are probably the best way to teach people

For me it's a reverse filter, the people I see doing the shaming, those the loudest most seemingly unhinged anti-AI folks - they get blocked / ignored and looked at as if they were the equiv of PETA activists.

Yes, I avoided using it for the longest but the voices that end up pushing me away these days are the antis who gatekeep like crazy, say other people are lazy, decry anything they don't like as "slop" etc. One side is moving forward and I'm sure they'll figure out the right mix over time. The group that just denigrates other people, I wonder how they will cope over the next few years.

> I wonder how they will cope over the next few years

Perhaps they will pool together enough widespread stories of issues with AI (drones, surveillance, misleading advertising, etc) and create a political movement to try to ban or regulate it; only to then have their movement co-opted by the people who just want to get paid.

Are you saying nobody should ever do anything with any tool if they cannot reasonably do it without the tool?

I mean personal blogs I agree the writing portion should be personal but does that go for everything? Maybe the author should take the emoji off his page(??!!)

> Public shaming and rejection are probably the best way to teach people it's wasting people's time and disrespectful

Public shaming is disrespectful. As is the self-righteous "teaching" of something that is ultimately a personal preference. You aren't being forced to read someone's personal blog.

This hasn’t been my wife’s experience at all. She manually writes all her content, but uses AI images as sizzlers for the scenes (fan fic for popular shows). Short of getting the actors together for a million bucks, she’d be hard pressed to make these scenes a reality. It’s also supposed to be aspirational, like “what if the tv cast got back together for another season hey?” I think the average person just doesn’t really notice.

She’s had to figure out where the line is though. She tried using ElevenLabs to dictate her scripts, thinking it’d make things more convenient for her fans, and that lost her subscribers because they seemingly disliked the more obvious use of AI.

AI generic blogspam "art" - as off-putting or worse than clipart or those weird flat characters that took over everything "corporate" a few years ago.

Actually using AI to make a shitpost image relevant to the text? Amusing, fun, worthwhile.

The bendy arm people

https://www.wired.com/story/corporate-memphis-design-tech/

Corporate Memphis! Thanks, I hate it.

Now there's AI Corporate Memphis.

> I think the average person just doesn’t really notice.

Yes, this is what's insulting about it, the assumption that it won't be noticed.

People also confuse “didn’t notice” with “didn’t confront me about”.

People noticed, they just didn’t want cause a conflict over it.

Yes I don't really bother to call out AI images or writing, I tend to just block the domain and not look back.

Maybe a few strikes if the content is otherwise substantial, but it's a very easy and reliable smell so far.

Manually written content with an AI teaser image is perfectly acceptable to most people, including technical audiences who dislike AI writing.

The teaser image isn’t the content. Most people consider it part of the web design or setting the tone for the article.

There’s a very vocal minority who will get turned off by AI anything right now, even when it’s not part of the main content. Some times a good blog post will land on the HN front page and it will collect complaints that the layout or web design looked too AI generated so those people didn’t read it. I don’t think it’s a big loss. Blogs have attracted a peanut gallery of picky critics forever.

> She’s had to figure out where the line is though. She tried using ElevenLabs to dictate her scripts, thinking it’d make things more convenient for her fans, and that lost her subscribers because they seemingly disliked the more obvious use of AI.

Using AI for the main content is a no-no. The actual content that they came for has to show signs of authenticity.

TFA is not about fanfic, which seems like a natural fit for illustration, but about technical blogs that really don't need it and are instead stuffed with distractingly fancy slop images.

Same, I tend to assume it is a sign that AI is used across the board.

But please keep it coming. I need to spend less time idly browsing and the growing preponderance of AI is helping push me in the direction I want to go. My time spent on “social media” has dropped significantly over recent months.

Part of the problem is many bloggers think an image is required for every post. This is reinforced by algorithms which favor such posts, and drives bloggers to the easiest/fastest sources of free images.

The "featured image" in Wordpress is what gets shown on the index page, in shared social posts, and other contexts. And certainly, the algorithms favor such posts over pure text posts, even if the text content in the spartan post is far better.

There are other factors at play, too. Many bloggers are lazy, some are severely time constrained, a few are fearful of legal repercussions associated with copying images off the Internet, and a majority are creatively or technically incapable of making their own images to include on a personal blog post.

The same goes for your business. Your AI “art” isn’t eye catching, at first glance it’s lazy. A longer glance it’s unsettling. We’re heading for a really boring period in graphic design.

For over a decade, people have been dropshipping slop on Etsy and apparently making money.

It’s not like stock photos of infeasibly attractive people eating ludicrously overstuffed burgers were ever interesting or believable.

Of course not. Bad marketing is bad marketing.

either way, looks like we're staring down the barrel of mediocrity - the AI isn't a fix. You can't really AI yourself outside of 'the average x/y/z' that these models generate. being specific with your prompting still leads to the most average version of said prompt...rinse repeat.

At least stock photos don't have the documented side effect of inducing nausea.

This is not surprising. Would you rather have handmade Italian shoes or factory made shoes sewn by a robot. Humans assign quality relative to themselves.

A chess game between to bots is boring, no matter how strong the bots are. Humans generally compare themselves to other humans and that's part of how they value.

I dunno which I’d rather have, but I do know the only kind I’ve ever bought is the factory-made kind. So if the aim is appealing to a wide audience, you’re arguing involving robots in the production process is key?

> A chess game between to bots is boring, no matter how ...

I'm thinking quite a few chess nerds could be interested - depending on how the game develops.

Some of the most fascinating games are between Stockfish and Leela. The quoted statement is grossly ignorant.

> Would you rather have handmade Italian shoes or factory made shoes sewn by a robot.

The latter, of course, which is why that's what I own. Aside from that, this is an absurd and grossly dishonest analogy -- that's not at all the choice here.

Some of the most fascinating games are between Stockfish and Leela. (e.g., https://www.youtube.com/watch?v=h_84B_VwayU -- this is narrated by National Master "Jerry", one of the most respected chess narrators on YouTube.) People who don't know this know nothing about chess. (And only immature clueless fish respond by challenging someone to a game. Lack of knowledge of chess was already established by their ridiculous claim. BTW, he has radically changed his response , which is now very different from his original claim. It's not my misunderstanding, it's his bad faith.)

Some of the most stupid things I have ever read are from those suffering from radical anti-AI hysteria, and it is becoming a primary means by which I judge people's intelligence. ("radical" is a key word here -- there are numerous legitimate reasons to be critical of numerous aspects of the AI boom.)

I won't respond further.

I think you misunderstood what I said. Let me rephrase.

Would you rather watch a chess tournament between two of the best human players in world, or between two of the best bots in the world?

I know that chess is deeper than this, its about the positions that are reached and what each game adds to the existing knowledge and in that case it doesn't matter if its a human or bot. I understand very well that there is allot a Grandmaster can learn from games between two strong bots.

But here I am talking about human nature on how we value things, if a human can do something that I can't do than I value that more, I know Usain Bolt is faster than me, I know he is extremly fast compared to myself and other humans.

I challenge you to a game of chess. Lets Go!

“AI-generated images discourage me from reading” is a statement that’s true for me too, albeit for different reasons than offered in this post. For me, it’s the same kind of thing as “reading the text of your slides out loud”, or repeating the contention in both the introduction and conclusion.

The point of repetition is to reinforce understanding by approaching the same place from different directions, so the reader can pin down your idea at the intersection of these various thrusts. Multiple arrows from roughly the same place going in roughly the same direction don’t constrain the expressed idea, instead they just make it fuzzier.

This has been a frequent problem on fiction writing on royal road. Author promises that they don't use AI to write and then proceed to use AI to generate the cover for the series. They might be telling the truth, but I am not going to spend the time to figure it out for myself. They are really shooting themselves in the foot. A shitty MS paint cover would be less off putting.

"I’d rather see a shitty Microsoft Paint drawing as opposed to some AI image."

Interesting because I'm seeing tons of YouTube channels using AI generated stick figure animations with AI voice overlays to trick people into thinking it's not AI generated.

The key there is you still notice it. The format is less important then the fact that there's enough signal there telling you "something is wrong".

Fair enough. I dislike word-expansion AI text so I get it. When I write on my blog, part of it is as memory like a diary for myself and I find that I can create these images that help remember the scene well. As much as I'd like people to read my blog, I get that they have preferences, and if many decide against then that's their prerogative.

Here are two examples with AI images from Nano Banana:

https://wiki.roshangeorge.dev/w/Blog/2026-04-20/Drops_of_Wat...

https://wiki.roshangeorge.dev/w/Blog/2026-07-23/Sixtless_in_...

I think I'll probably keep doing what I'm doing there.

The painting style does make these look less obvious than the default, so that helps. I think this is the best generated art I've seen so far.

Although if it's for a mental diary maybe you could get a nice camera for it?

that's pretty nice illustration work, I'd agree, keep doing it :D

I thought Nano Banana did a great job! I used a trick I saw someone describe elsewhere where I inserted the annotations on the photo / google street view of the scene. It seems like the model does a better job of instruction compliance when you annotate on the image itself rather than provide text. So an arrow to place the penny farthing guy and so on.

it seems kind of perverse to to generate an image of a scene (fundamentally not derived from how it actually looked) to remember how it looked; you even immediately point out a way that it's inaccurate ("They looked younger and in better shape but you get the picture.")

It's pretty neat, actually. If you use Gemini's Nano Banana with a photo reference, it's pretty good at keeping most of the things that matter. I walk down 4th Street daily and that's pretty much what it looks like near Berry St.

And the thing I'm trying to remember is not, say, her athleisure or his white jacket. I'm trying to remember a couple with a dog with the man carrying a bag and a garbage grabber.

This notion of precision is actually an interesting thing to me. It's a question of what I'm trying to memorialize. Their clothes or their fitness or the colour of their dog doesn't matter so much.

You know, like some people object to statues of John Muir because he said some things like calling indigenous people "dirty and irregular" and stuff like that. But I think statues don't memorialize the man, they memorialize the symbolic man. The actual John Muir is dead, and we have his memetic descendant John Muir The Image. John Muir the image is a chalk drawing of John Muir the man, and the fact that the latter liked a dog, or hated boiled potatoes, or loved a woman, or whatever is irrelevant to me. I have taken this person, and flattened him on a canvas, and painted his acts on them as a guideline for myself. The rest of him is gone. And I like it that way.

I've had this as a draft post on the blog for a while so if you don't quite get what I'm saying about this accepted imprecision, it's probably because I haven't yet found a way to write this idea out in a clearer way.

I think you would find learning to draw extremely rewarding

Knowing how to, perhaps. Learning how to, sadly not. But my wife is an artist / creative director so there’s some adjacency there.

For my wife's birthday, I generated a fantastical image of our family (me, her, daughter, two cats) because staging such a thing IRL would have been difficult. It took me several hours to take some nice photos of my cats and daughter and then composite them and then generate an image i liked. Then I sent it to a framer and got it framed.

She loved it!

I think there's a time and a place for AI generated images that actually require care to generate a good result. I guess to use a cliché, its the thought that counts.

Haha, it's all in how you use the tool, eh? My wife actually used an early Dall E to iterate on the design for my wedding sherwani. I loved the end result too.

Picture here: https://x.com/arjie/status/1855328068883353665?s=20

So much this, blogs add an image for the sake of having an image. At this point it would be much better to just have a random dog or a cat pic, it is as relevant to your blog post as your random generated thing, but I'd prefer seeing a random happy pup.

And if you generated a pic I suppose you also generated your post, so I'm way less likely to pay attention to it.

if cybersecurity blogs replaced AI hacker images with dog pics the world would be a better place.

I have a compromise: An AI hacker image using a picture of your own dog.

Exactly, I totally forgot there even are existing apis that return open source dog and cat pictures

If that dog is wearing Matrix style sunglasses... maybe a cute dog hoodie, I'm even more interested.

I've recently started doing crude drawings on my reMarkable when needing something like explainer diagrams. It looks like a toddler did it, but, you know what? It works. I just stopped feeling weird about it.

Pen-type input systems are pretty cheap these days (no need to go reMarkable, I just use it because I have one.) It's not a super frequent need but having something of that type in general I'd consider valuable enough to spend the money.

If you're writing a personal blog (as opposed to a company blog), you should not care what people think.

A blog is an expression of your own taste. Write about things that interest you. Write in a style that you like. Add whatever images you want. And don't let some random stranger on the internet influence your blog. Otherwise, it's not your blog anymore.

Couldn’t agree more. Then express yourself freely. Don't let stupid grammar limit you. Otherwise it's not your blog anymore.

Not sure if this is sarcasm or not, but if someone wrote a blog post that said, "Bad Grammar Discourages Me from Reading Your Blog" I would have the same response: Don't write your blog to please other people.

I write a blog and have always avoided AI images for exactly the reasons cited here. AI images feel like a more cynical version of stock imagery, for some reason. I think it's actually too bad because in a lot of cases AI images would actually be really great to include and a lot more evocative... but there's just no way to do it that doesn't inspire revulsion.

It just makes me wonder what else in your content is AI-generated. Just sows a little seed of distrust.

The goal of someone putting actual ideas out into the world should be to establish trust, not create questions about whether the ideas are theirs or plagiarized from ChatGPT.

For me the breaking point for images in texts was stock photos.

Before that, when there was a photo in the article it was a frame from a footage from the event or the photo, or the schematic of the device the article was about or something. Something relevant to that particular piece of information.

But then everything became click-bait factory illustrated by images that fit the mood.

AI images don't make it worse. Don't make it better. But they might if they get better at relevant infographics. For now they aren't good enough because they rely on a human for understanding of what are they to depic and humans often don't care enough.

There is no "before that", not on the internet. Stock photography as a business is more than a century old, and we were using it on the internet as soon as you could put an IMG tag in a page.

Lazy use of stock imagery is a problem.

Your one-short-paragraph blog post with a shallow personal preference discourages me from reading your blog.

There was zero depth to it.

Human slop is slop too.

Human slop is a much better kind. There were still humans involved.

I think that people upvote these kind of posts for the sake of seeing if can generate any interesting conversation in the comments.

yeah, and half of their blog is rate limited for some odd reason due to traffic

that seems lazier than whatever they're gatekeeping about

they even tried to get in front of that criticism

> at least you know for a fact you’re getting the thoughts of a real human being

No, we don't. For all we know this whole blog is an agent larping as a neverLLMer. Your stance says nothing about who wrote what.

Just as using images freely generated by ChatGPT for a PERSONAL blog post because the author liked it but couldn't express it themselves doesn't tell you anything about the quality of the post. It tells you they're not an artist. It tells you that they prefer some imagery over blank space.

The rest is you and your projections. And that's okay too. Jog on.

I've developed an internal heuristic that tells me people who think an AI image to fill some blank space in any way improves anything aren't saying anything of value. There's a strong likelihood they're just trying to promote some product or scam. I came to this conclusion after seeing 100% of sites with AI images plastered on them being scams and BS with no actual value.

Meanwhile, a blog that's basic HTML and text and looks like it's fresh from 1998 is generally high quality info. People who are willing to write like that and publicly share it have confidence in what they're writing. And if that blog spreads, it's because the content is good. Not because it was eye catching at a glance.

How does your heuristic handle your qualitative and quantitative words here? If you land on a blog post with a single, pertinent illustration (from a LLM), do you assume the post, site and author are all robot trash and run away? Investigate the other posts before reading further?

The point of my caps, above was that I'm writing for me. If I think my post is improved by me prompting a LLM for an illustration, that's up to me. I'm not writing for your heuristic, I'm trying to express myself every way I can. The time where you find individuals untainted or unaided by these tools is over.

I'd try to just content by it's content rather than it's illustrations.

I don't assume it's trash. I know with absolute certainty. It's the same as seeing someone posting with one of those old NFT monkey avatars. You know they have nothing of value to say, and they're most often going to try selling you something of no value.

Fair. The same is true of stock images and a lot of other visuals. Some writers just include images that are relevant, but a lot of them read some study indicating that imagery generated X% more page views and y% longer reading times and just developed the habit of sticking something on every post. If you think those studies are valid and are still inclined to do this, consider mining Wikipedia or some other site for obscure but interesting things and including a thoughtful caption, as well as a link to the full resolution image.

I have the same reaction with AI generated YouTube thumbnails. Something in the back of my mind immediately goes "it's probably low quality content" and I'm less likely to click.

I have the same reaction with the "YouTube face" thumbnails.

If you're writing about something which exists in reality, then it is possible to take a photo of it. If you're writing about something which doesn't exist in reality, then it shouldn't be accompanied by any image.

Stock photos or AI generated images are always wrong. Real photos, no matter how bland, convey information from reality and are always right.

I find them very interesting and they make me more likely to remember your blog. I was reading that Gas Town post yesterday and despite all the backlash about the whimsical world and the ai images being sloppy and wrong, they still stuck in my brain more than anything else I’ve read this year.

The author say they’d prefer Microsoft paint images but actually I don’t think this is true of most audiences. People have the ability to make terrible paint images for decades and they’ve never caught anyone’s attention in a positive way, it’s not going to start now.

OP’s blog of white text on black background will be forgotten quickly. If I visit it again, I will have no recollection of ever having visited. I’m not saying they need to use AI, but to oppose AI with the opposite —- minimal design or even intentionally crappy design — will not have the intended effect of communicating “this is quality human-authored content”, it will just be forgettable.

> People have the ability to make terrible paint images for decades and they’ve never caught anyone’s attention in a positive way

Citation needed for this one. It could be because people don't do it very much? I think a very good piece of writing coupled with a crappy but intentional MS Paint piece could be very attention catching and memorable.

> OP’s blog of white text on black background will be forgotten quickly.

I actually get annoyed with the tech crowd always complaining at seemingly the slightest bit of flair or fun added to any blog. To each their own, of course. I've seen some really slick minimal blogs but I do like it when layouts gets creative--gawdy, even!--, so long as I can read the text. And you can always use reader mode.

On the other hand, plenty of art can lean into being intentionally brash and be all the more memorable for it. The phrase "bad MS paint blog" immediately makes me think of Hyperbole and a Half[1] where the style feeds into the zany ADHD vibe of the blog.

Often, you find ugliness and rule breaking as a backlash to an overly polished status quo. AM radio rock gave way to punk, hair metal gave way to grunge, and EDM gave way to hyperpop. The most iconic album cover of 2024 was the word "brat" badly rendered on a plain, garish background.[2]

[1] https://hyperboleandahalf.blogspot.com/2010/05/sneaky-hate-s...

[2] https://en.wikipedia.org/wiki/Brat_(album)

Yeah that’s cool as an intentional style. Here’s my favorite example, and the style definitely caused it to be sticky and viral:

https://m.youtube.com/watch?v=kCpjgl2baLs

If the blog is otherwise not about being amusing or edgy, then the MS Paint is going to send a different vibe. In your examples the paint sets the vibe, which I think is what they were going for.

Sweet Bro & Hella Jeff is still one of the worst greatest of all time

http://www.mspaintadventures.com/sweetbroandhellajeff/?cid=0...

From [0] https://nelson.cloud/about/#skills

> Other: Claude Code

Then the author said:

> I know there are plenty of things you can roast my blog for but at least you know for a fact you’re getting the thoughts of a real human being and not some LLM.

So it is OK for you to use AI generation tools like Claude Code and likely using it in your blog or using it to generate code, but it isn't OK for you to see someone else using AI to generate images?

With the above already stated, that means the author cannot guarantee that not a single use of AI was used in their blog at all given that we know they use Claude.

Why preach fake principles with virtue signalling and then break your own standards in your own blog post?

Someone who needed work listed a skill in their CV which many employers demand. Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize.[1]

[1] https://news.ycombinator.com/newsguidelines.html

They did that all by themselves.

> Someone who needed work listed a skill in their CV which many employers demand.

Then don't preach that you don't like AI yet you use it at work.

I don’t have a problem with ai generated media in blogs at all. Nobody is going to take the time to hand draw a piece of art for a blog that takes them less than an hour to write. Pictures speak a thousand words and serve as a great intro to a blog post. Embrace the future.

Hurray for an intelligent comment. It's alarming how much idiocy results from this radical anti-AI hysteria. e.g., someone here said that a set of slides "packed with information" was "completely worthless" because it lacked "an actual point of view" from "actual people".

> Pictures speak a thousand words and serve as a great intro to a blog post.

If pictures are so important, why would "nobody... take the time to hand draw a piece of art" or find a one of the many free stock photos? When I was actively blogging, I would doodle quick images to include all the time for this precise reason.

Because it took too much time before and now it’s essentially instantaneous and free?

My point is that your statement:

- Doesn't come across as consistent: You said pictures "serve as a great intro," making them a valuable addition, but simultaneously "nobody" is going to take the time to draw them? Why wouldn't an author take the time if pictures are such a great addition to their piece?

- It's also not true: you said "nobody" does this. I'm telling you I have. Here's another example[0] in this very thread of somebody who does the thing you're claiming "nobody" does.

Aside from that, the amount of time it took to write the post isn't even necessarily relevant. I think for many it's about trying to fully express an idea, presentation, enjoyment, clarity, compulsion, polish, [display in search results], or 100 other things.

People did it before and they do it now. Everybody doesn't do it, but it's not true that "nobody" does it. That could be part of why people's brains are honing in on it as a heuristic.

- [0] https://news.ycombinator.com/item?id=49169003

I'd rather read a blog post that includes relevant and decently made illustrations of the topic at hand (explanatory graphics etc) rather than just a wall of text. Makes no difference if these are drawn by hand or using X software or an LLM. But that's just me.

This is the weirdest thing—-I didn’t think anyone was paying attention to graphics and images anymore to care who produced them.

My personal signifier of fake content is that voice on videos. It’s most found in ads about foods. Or says, “the thing they won’t tell you…”.

I associate it with foreign language speakers making ads or content spam on YouTube. Or farming views with voice overs about WWII battles.

I recently heard it on a YouTube viddy about ICE https://youtube.com/shorts/56btguFIhP8

This voice seems fake to me.

Same with YouTube thumbnails for me. If it is obviously some exaggerated image nothing like what is in the video then i ask youtube to not recommend them. I should just stick to subscriptions but fishing gets me sometimes.

Some content creators are too focussed on their work that they might not have noticed the debate going on online about AI-generated images and all they wanted was a cool image for their content. They might not have known that they walked into a hornet's nest.... but they will find out soon enough. Should they have gone down the alternative route, where they spend time and money hiring someone to do it... who will probably use AI anyways. Telling someone not to use AI is like telling someone to not drive and instead take a horse and buggy (circa 1920).

I seriously doubt there is anyone making content for the internet who is not acutely aware of the AI debate.

Totally false equivalence.

It’s not about using new technology, it’s about the signal of effort.

Blind copy-pasting an AI image anywhere is a poor signal of effort, because it was a deliberate decision by the author… “I didn’t do the work to make this look professional, but my hope is that people THINK it looks professional.”

Genre images are frequently used to give a post an aura, as a kind of decoration and serves no other purpose than that. Then there are the images that does not quite fit in, that gives that AI ick. Few, if anyone, like them. They demonstrate lack of effort and demand your attention anyway. An imbalance in the force most can immediately sense. There needs to be some originality, some original thought or expression, a creative threshold needs to be passed. Which tools were used to create an image is not important, it is the who and why that is.

A while back I had the same feeling and hypothesised about a "nutriscore" for AI content in media:

https://disclosure.launchbowl.com/ https://launchbowl.com/ai_nutriscore

I think AI images can enhance a blog article as long as the reader is fully aware and trusting that it's just the auxiliary material to the blog has been generated, not the main body of value.

If you use images on your blog, even if you are scrupulous about sourcing them, purchase licenses, or go by the same low-resolution "fair use" rules that Wikipedia does, you will get several letters a year from law firms claiming they represent a client who owns the right to that photo.

Rather than wasting time sorting all these out when the images are there mostly to break up a wall of text, some blogs simply use a few AI images. This way if those letters come, they can simply put them in the bin.

I don't object to this use of AI images at all.

These are all symptoms of the adjustment period following a disruptive technology.

Eventually it'll be perfectly normal to use AI-generated content (and use AI to summarize it), and nobody will bat an eye.

Yes. Just like, shortly after the web, it became normal for 99% of content to be moneygrubbing adspam SEOslop. Normal doesn't mean good.

Eventually well formatted blogs won't be a symbol of effort and no one will care.

Just like published books were considered a symbol of effort that no one would have considered a blog as being anywhere near. When was the last time you looked up a book to find an answer to your question? That USED to be the normal.

Then we just googled for a link to Stack Overflow. And the oldtimers waxed lyrical on Slashdot, bemoaning the loss of craftsmanship, and how it was too easy nowadays - no effort.

What has been will be again, what has been done will be done again; there is nothing new under the sun.

The likely next normal will be agent-scraped content that it summarizes and tailors to its human. Slop, lacking real content, will be dutifully skipped by the agent, and the human won't care.

Meanwhile, armies of advertisers will cry foul, and scramble for a new way to get in front of human eyeballs.

And for those who want to continue reading blogs for pleasure, it's one agent pre-screen away.

Tempora mutantur, nos et mutamur in illis.

Both blogs and books are mostly spam now, anyway.

Just like Taboola. They are everywhere now.

I think by "disruptive" you mean "annoying"

There is negative value in wasting vertical space to put an irrelevant prompt output in an article. It makes me question the sanity of whoever is doing this stuff. Doubly so if they actually think they dumped a bunch of AI-generated text on the internet just for people to AI-summarize it. Like... what are they DOING?

The normalization process for social media involved it stabilizing in a place where a huge number of people know it’s terrible for you.

Agree but how many times are the same articles and opinions reaching the top. It feels like we’re in a loop and it’s boring me out reading the same thing over and over and over again.

IMHO AI-generated images are not the problem; The low effort people put in to generating them and selecting which image to go with is the problem.

Spend some time on it. Find your own style.

Most of the people that do this don’t have the basic human decency to start with a clear message like “All or some of this content is AI generated.”

If you say that, and I stay engaged with your content, that’s on me. But if I keep reading, and I find out later you’ve done this to me, then I probably can’t take you seriously anymore. I don’t care of it’s one picture in a ten thousand word essay.

I don't think the internet as we know it is going to survive generative AI.

Before, the simple existence of the content of some basic level of quality was a strong signal of effort, which corresponds strongly to value. If it's there, it's worth reading.

Now, the existence of content of a basic level of quality signals nothing. It might have some value, but it's much more likely it's useless AI slop.

This principle -- if it's there it's probably worth taking a look at -- undergirded so much of system of creation, discovery, and sharing that made the internet special -- blogs, open-source, posts, comments.

It allowed the gatekeepers, middlemen, guided channels, and tolls to be removed from the creation, enjoyment, sharing, and consumption of content.

I'm worried we're going to have to go back, just to get away from the avalanche of crap people are using AI to produce at crazy rate.

Ok? That's nice.

As someone who adds AI generated images on blog entries (a single context-appropriate genAI image that is used for OpenGraph sharing), I just do not give a shit if some people have opinions like this. Jog on?

The vast majority of bloggers in 2026 do so for their own edification, to their own preferences and choices. Whining about what they do is decided a beggars being choosers thing.

100% agree. If I see an AI-generated image in a post, there’s almost no chance I’ll read it. The same goes for YouTube thumbnails.

This is why I started creating all illustrations on my blog by hand. Maybe they are crappy but they are my crappy drawings.

When I was working on a post recently, I rediscovered that it is actually fun to draw. Much more fun than debating with an AI tool to generate an image for me. And given the time I already spent on the subject matter, the effort was not that much more. I can recommend drawing on paper. You can correct mistakes after scanning.

"I’d rather see a shitty Microsoft Paint drawing as opposed to some AI image."

The problem is, there are also AI generated MS Paint drawings that you can't tell is AI. We are only noticing the AI images that are in an 'AI' style. The AI tools allow you to make images you can't tell is AI.

Is this about technical blogs where the image is either of technical nature or something else?

I'd think in the former case it doesn't matter and it's mostly the text I care about to be interesting and personal. I don't read non technical blogs so I don't know what if ai images are a turnoff.

Not all of us can afford to hire illustrators for every blog entry, and while an MS Paint squiggle might suit this blog's author's tastes, I can sympathize with anyone who might prefer something more sophisticated.

I don't see what the problem is with AI images, if some thought has been put into the image itself and if it's relevant in some way. I'd far rather see something playful and on-topic than the generic stock images littering blogs previously or some random example of the writer's photography with little connection to the article.

But really, whatever I like or don't like is besides the point. The blog using AI images isn't my blog. And if I was one of the blog authors who've sent this author into full prig-mode, I'd say to him, "No worries, Nelson. Have a happy life." I'd wager the blog author's own life wouldn't be any happier for bowing to the sensibilities of the Nelsons of this world.

AI images are not more sophisticated, they are less sophisticated

I think the issue is we all have read a bit into a AI article before realizing it is one and realize foolishly that you have been reading drivel.

The original prompts might have had something interesting to say, but it is so layered with filler and platitudes that it is just a waste.

After being burned enough times, we hunt around for signals that an article is about to waste our time.

Not all articles with AI images are LLM written, but almost all LLM written articles have AI images.

The author is telling you a hueristic is that people are using, and for this person here, I am totally doing it.

A picture of a napkin doodle at this point is superior.

Some people are discovering the trick of “humanwashing” AI images by tracing over them by hand in standard drawing programs and applying their “taste” to produce something that looks like a human created it in the first place. What do we think of this?

Shout out to Pixabay. I most often use them of cc0 images, though recently I had to start using their filter to exclude AI generated. Never understood the need to upload AI generated pictures there when anyone can just generate them on their own.

That may be your preference, but it's quite contrary to mine.

I have successfully used AI to generate generates "concept art" and logos for projects, presentations, and publications. It seems to make the outcomes more approachable, more understandable, and certainly more aesthetically pleasing.

It is not a matter of replacing human effort or skill, and certainly not a matter of adding cheap ornamentation. If AI weren't available, I usually wouldn't have the time, money, energy, or will to pursue those illustrations. I say this as someone who has historically spent what seems like a vast amount of time, energy, and money sourcing art. In a former life as as publisher I even hired full-time artists and insisted on custom artwork. But my current work no longer has that luxury. I still want the best possible looks, and the best possible outcomes.

See other top HN article rn about "expertise is essential for using LLMs". Nobody is debating whether illustrators, graphic designers or marketers are going to use AI. In addition to being a thoroughly sailed ship, the sheer demand for constant content and image generation makes it a job skill at this point.

TFA is about blogs written by people with non-graphic expertise, who think it's great to, e.g., punctuate every other point with a way-too-detailed picture of a panda bear coder orchestrating fleets of squirrel agents. (Yes I'm looking at you Steve Yegge.) Nobody needs to be abused with this crap, and it all has the same overpolished overdetailed look, the "it's too good to be drawn by the author but is obviously NOT drawn by anybody with any visual sense".

It's not just blogs -- bands are now making these ridiculous "pro" fliers, food carts have stupid pictures of their owners on them -- but in blogs that ostensibly are written by humans for humans, AI images are worse than distracting and a very negative signal.

I find AI generated text way worse than images. I get a migraine from reading it.

The AI image is a good quick signal that the text is likely AI generated without even reading it. I say let's keep them around as a shibboleth.

And then there's people that use the AI-generated images to intentionally drive away the people that don't like the pics and AI so as to cultivate a reader base of like-minded people who are ok with using AI

I really don't get the AI hatred. I have seen impressive work created with AI and really shitty work created with AI.

The problem is not AI. The problem is the person using the AI.

Couldn't agree more - down to the shitty ms paint recommendation

Finally. Same effect with stock imagery. I like using a creative imagery. Hacky looking tuis like sl or pipes.sh. They have some eyecandy and make my blogs have a uniform style.

I'd say I don't like images that feel too AIsh and have errors/quirks. Still, some AI images are decent, depends on the person and his taste.

I used Claude to generate technical SVGs (which are verifiable correct) because I wasn't going to take the time to draw the grid/arrows/etc.

Only if it has that AI look. What you don’t realize is that you can generate images that don’t have “the look” rather easily if you have an eye for it.

The problem with illustrations is that AI fills in so much nonsense, it is difficult to tell of the picture’s thousand words which ones were intended.

AI images are low-res information masquerading as high-res.

It's a lie. It tricks the reader looking for deeper meaning.

Though hastily chosen non-AI images may do this too.

Eh. I tend to use stock photo stuff. Even easier than AI (and most stockhouses are now using AI, themselves). If it's important to the story, I'll generally generate it, myself (I tend to use OmniGraffle a lot). I'm quite capable of making my own images.

The biggest thing, for me, is provenance. I sometimes get amazed at how people just grab any image they want, and slap it into their blog. There's a whole class of scammer, that goes around spamming fake "we're gonna sue you" letters to sites (I have gotten them, for images that I know are completely clear of obligation). If the image really was ripped off, then they won't be scammers, sending the emails.

I'd rather someone use an AI-generated image; especially if it's just "window dressing," than they put themselves into legal jeopardy, by pirating images.

My my knee jerk reaction is to agree, I really hate the obvious looking AI images, though I suspect ultimately it comes down to having good taste.

As much as I like to be on the anti AI bandwagon, for which there are plenty of good reasons. Some people use the tool (like any tool) well, to compliment their work.

Eg. this author uses great images to accompany their articles (except for one which is obviously AI), its hard to know if they are real or not but they support the text. https://tuhat.net/@jesusrrodriguez/

Regarding businesses, I think the use of bland stock images made most corporate sites awful years ago. AI is just accelerating and making it easier for the rest of us to create the lazy heartless "content" that has been so common from the corporate world.

I'd prefer no images to AI-generated ones.

This seems temporary. AI generated images will become the norm for many applications. Today we wonder whether a model in an advertisement is real or generated, 5 years from know this wont even be an interesting thing to consider.

In many contexts, images are meant to illustrate a point. The human becomes the gatekeeper of generated content. The question becomes: does this picture illustrate the point I want to make. If so, you are adding your human approved stamp to the genrated output.

100%. if I see AI slop images or videos I just click off, it's become almost automatic for me now.

I'm not gonna spend my mental bandwidth on figuring out if the thing is worth reading/whatever and that somehow the author just didn't realize that it's a net negative

I think the discouragement started long before AI, probably before GUI, based on the terminal-like website you host.

I wonder if this sentiment will endure, or if younger generations that grow up around AI will be less discerning?

Have you seen Midjourney 8.2?

Pretty soon, you will not be able to tell.

What you really mean is "obvious AI-generated images".

Last time I checked, their landing page looked like, I imagine, an escort agency catalog would. And why would I go to a cesspit if I already disturbed enough by that "art" screaming at me from every corner of Internet?

The main problem is, those AI generated images are only so jarring because they're so obvious.

As time goes on they'll be harder to spot. We're going to need a new way of determining quality as such obvious signals have a limited shelf-life.

Not that AI generated images are always bad. They are perfectly acceptable if used for (not an exclusive list): small separators between sections in large blog posts, or other decorative features; used to illustrate things for which no real image exists (as long as it's labelled as an impression/reconstruction/etc.). They're only really a negative signal if they're used dishonestly, to claim something existed when it didn't, etc.

Indeed - instead of pointless hero images, now we have pointless ai-generated hero images.

I use them, and the reason is simple - they attract users. All these directions to not use AI here, not use it there, not use it anywhere, they're all basically missing the point, which is what someone was trying to do in the first place.

Without AI images, maybe some stock images? Or no image? No image is statistically shown to get less engagement on platforms that show images, so using no image is conceding to someone with a better marketing department. Stock images can avoid it being AI, but can be quite bad unless I have a lot of time. Make my own? Well, now I need a new skill, and instead of expressing my idea through one medium, I have to be skilled in two?

And, next, don't use AI as an editor right? This is similar to the images trade-off. The writing is better when you have an editor. Sure, a professional editor is better (for now), but professionals are the ones with access to those.

I just want to express an idea. Why does it have to be so difficult? Well, of course, one, because expressing yourself is hard on it's own, but two because of competition. I have no problem taking responsibility for the first, but these rules around the second are annoying. The anti-AI rules are just the tip of the iceberg here, but I think some awareness that appropriate use of AI is an attempt to overcome what were often dumb rules and biases.

I totally get how the non-appropriate AI use is creates anger, and how that bleeds over to the appropriate use. But you know, I wish there was something similar about all of the non-appropriate things that are obstacles to communicating that pre-existed before AI.

In the 80's, many, many people looked down on electronic and synthesized music because it wasn't real music played with real instruments.

And now it's just another instrument and has whole genres of music dedicated to it. C64 SID chips are sought after for 8-bit music creation because it has a unique aesthetic. Same with TR-808 drum machines.

The AI complaints feel a lot like that.

I'm betting in about 15-20 years there's going to be a revival of terrible AI images and videos because people are nostalgic for 2024 when AI images were obvious and sit firmly in an uncanny valley.

And I think AI will inspire a massive artistic revolution, both because it becomes just another tool in the toolbox that makes creation easier like drum machines and synths, AND it takes over the pop-music space previously occupied by boy bands and pop queens - generic rehashes of inoffensive crap that so saturates the market entire counter-cultures develop. But someone will use it to make something uniquely creative and new. Rap and hip-hop wouldn't exist without cheap drum machines.

In other words, you're old enough to start disliking new stuff, but not old enough to have seen this happen multiple times and realize that this is just another generational cycle. The younger generations will absorb AI tools and continue to do what humans have always done, which is be creative with the tools at their disposal.

That’s very rich coming from a person that uses monospaced font for a blog.

Belongs on Luddite News

Cheap AI image generation is a miracle and this article is dumb af

The moment I see anything LLM-generated it's an instant close tab.

Me too. Same goes watching videos that have AI generated thumbnails.

I use AI-themed images for the opener to my blog. It's typically a steampunk robot image that is not related to the topic of the post.

The actual content is written by me. If there are any code samples that were AI-generated, I'll call them out as such.

> AI-themed images for the opener

Yes, this is the exact kind of AI image usage that discourages me (and presumably this post's author) from reading the rest of it.

that is exactly the kind of thing that makes me close the tab.

like why?

to me it just means "I'm an author that likes/makes AI slop and this article is probably half if not complete AI slop or not worth reading"

I don't see how that would add any value or boost page visits

I find it entertaining. I'm doing it for me, not to boost any metric.

I am my target audience.

I think you should learn from the reaction of most people in this thread and avoid it (if your audience is mostly technical people) for the following reason.

A. If it's unrelated, it doesn't add anything to the post except aesthetics. B. On the aesthetics dimension: I don't think people are very impressed by the AI aesthetic, and tend to be annoyed by it. C. It seems that most people now assume that AI-generated image == AI-generated text.

I personally do not care about AI images or not, but it seems that a portion of people do.

Alternatively, consider that the people who are so against this are posting here with great conviction, and everyone else who isn't against it just reading the blogs rather than writing on Hacker News.

I'm personally growing an hatred for the genetic fallacy...

What if my blog is mostly about AI generated images? :P

Would photography discourage OP from reading a blog? Should I hand-paint my imagery? If I am the type of person that would use a camera, then maybe I'd be the type of person to hire a ghostwriter.

AI images are not photography.

I mean, it's a spectrum, surely...

What about stock photography, IMO as much as I personally hate "overcooked" AI-generated imagery, it doesn't annoy me that much more than stock photography does on blogs.

I do however use it as a gauge of "if the images are generated by AI, how much of the text content is?" though, so...

Yes, stock photography is also lame, but at least their creators gave permission. It's my dream that "AI" is going to make us look back and reconsider some of the pre-era slop we've accepted in the name of convenience. I doubt it, though.

It's not a specturm. It's quite simple; AI jpegs are not photography, because no photons were involved in the AI halucinations.

Yet the kids can't stop doing either... Which was the OP's point.

"the kids" it's mostly boomers falling for this and being obsessed with it

"the kids" are the ones who can spot AI slop pretty easily from what I know

Charles Baudelaire argued photography could not be considered a fine art because it merely produced mechanical replicas of reality without input from human imagination. He called photography "art's most mortal enemy" and the "refuge of all failed painters" who lacked artistic talent and training.

And if he's to be believed in 1859, photography isn't art.

But today, it's widely accepted that it is. And in all likelihood, you think it is too.

Why is that? And if you have an answer, you'd have to argue why AI generated images wouldn't follow the same trajectory.

Putting aside your desire to debate whether photography is art or not.

AI jpegs are not photography, because no photons were involved in the AI halucinations.

There's no debate over whether photography is art or not, and that's the point. Photons are neither here nor there.

The AI debate will be settled the same way and for the same reasons.

What does that even mean??

Photography is a form of art and requires human work and creativity, it isn't some clanker slop.

Okay but that’s exactly what the painters said about photography when it was invented. They said it was more chemistry than art because of the development process.

  Photography couldn’t qualify as an art in its own right, the explanation went, because it lacked “something beyond mere mechanism at the bottom of it.

https://daily.jstor.org/when-photography-was-not-art/

I don’t see how blanket labeling all images made with AI as slop is any different from people rejecting photographs because they were made with chemicals and lenses.

In 100 years the same article will be written about AI art, and will probably feature the phrase “clanker slop”.

Can you show me one AI image with artistic value?

I can show an AI image with artistic value.

But, to you? Apparently not.

Do it then

My comment was referring to the fact that art is in the eye of the beholder. Given your biases, nobody could show you an AI image that you would say is art.

Yeah if you want recommendation of things I like, drop a contact and I'll send you some. But I don't really want to get into a public debate about taste in art, because that's all posting anything would come down to.

It's a losing battle, because the bias is so strong you can post an actual Monet but say it was generated by AI, and people will come out to tell you how it's clearly not art:

https://www.threads.com/@merzmensch_kosmopol/post/DYUaBV2CbE...

Which is funny because that's how Monet and the Impressionists were originally received by critics. The comment about the light rendering being wrong, so clearly it couldn't be Monet is especially funny given the origin of why they are called Impressionists:

  Le Havre, he called it Impression. The term, springing from the jargon used by painters, had been an artistic watchword since the middle of the century, reflecting the growing interest in capturing the atmosphere of a scene, or impression, rather than giving a painstaking description of nature

https://www.marmottan.fr/en/notice/4014/

So give the above it's really futile to debate over whether any particular image is art. But yeah, drop a contact if you want recs!

Just post a link bro it's not rocket science.

I told you why I won't, I'm not getting drawn into an argument about aesthetics. What I personally think has artistic value and whether or not you agree doesn't matter at all.

These artists are pretty cool:

- https://andreaciulu.com/ - https://www.rooperainisto.com/

You might feel differently, but I feel intention and effort being communicated here.

Exactly my point. People simply weren't alive in the beginning of a medium, so they think how it is now is how it was always so. For the first 20 to 30 years of film, people disregarded it as art.

And somehow, it's a leap of logic that just because the AI image is generated that the text is also. It's as silly as asserting because someone used a camera, they hired a ghostwriter.

Your quick take lacks perspective. Photography wasn't always viewed to be art. Critics called photography "slop" in its infancy for a long while. See Charles Baudelaire for his 1800's take on photography:

> As the photographic industry was the refuge of every would-be painter, every painter too ill-endowed or too lazy to complete his studies, this universal infatuation bore not only the mark of a blindness, an imbecility, but had also the air of a vengeance. I do not believe, or at least I do not wish to believe, in the absolute success of such a brutish conspiracy, in which, as in all others, one finds both fools and knaves; but I am convinced that the ill-applied developments of photography, like all other purely material developments of progress, have contributed much to the impoverishment of the French artistic genius, which is already so scarce.

Harsh. It wasn't a fringe view either. I encourage you to read how much creatives despised photography.

This argument was transferred to film when it first came out as well. It took a long while and a couple court cases before it was accepted that it took editorial choice for a photograph and film to be good. When it was accepted that to be the case, despite the aspects of mechanical reproduction, it was also accepted into the pantheon of fine art.

So it will be for AI generated images and video. So far, I've seen nothing inherent in the properties of AI generated media as a medium that obviates a human editorial eye. Even if AI could exercise comparable judgement, we will like an image because we want the curation of a specific person.

In Grant Sanderson's interview with Dwarkesh Patel about the future role of mathematicians (36:25), he says:

> One interesting take that I've heard about what mathematicians will end up being is that it's actually more analogous to art museum curators than anything else. The AI solved the thing, so the art exists. They even know how to explain it really well. But you still want someone to help you navigate this nearly infinite space of what ideas are worth engaging with. Even if AIs were in some sense better at that, I think we would always still prefer a human that we had a relationship with, because the way we get motivated to be interested in things is a social phenomenon. If you have some specific technology you're trying to build, that might be different. But the people listening to this podcast trust your curation on what's an interesting topic in the first place. It's not that they're landing here because whatever your next topic is, that's what they wanted to understand in a prior sense. They're trusting you as a curator. So my role, and arguably that of other mathematicians, might actually just shift subtly into that curation direction of what ideas are worth pursuing. That's a lot of my job right now.

Hence, I think there will be a role for AI generated images and video to be elevated into the fine art in the next couple of decades, and we'll see them in museums and exhibitions--as well more mundane uses like what's sitting on your phone right now--just like how photographs are used today.

> Hence, I think there will be a role for AI generated images and video to be elevated into the fine art in the next couple of decades, and we'll see them in museums and exhibitions

Show me one example of AI "art"

Would you be able to pick out an example of "good photography as art" in the first 6 years of photography's existence, given generative media has been around for about 6 years?

Does the lack of an existence proof for fine art photography in the first six years mean that photography will never be fine art? Obviously not for us living in the future.

You'll need to propose a different, more honest test.

> Would you be able to pick out an example of "good photography as art" in the first 6 years of photography's existence

"Self‐Portrait as a Drowned Man" is thought to be the first staged photograph and it carries a powerful artistic and political message.

Literally everyone in the world with access to a browser can prompt chatgpt for an image. Billions of AI images have been generated yet you can't point literally to a single one.

Illiterate hunter gatherers in the Paleolithic with a life expectancy of mid-20s scrounged what few pigments they could muster and left their hands' imprints on stone. Even in this age of globalization, immediate access to information, and the sum of human knowledge on your pocket, having stood in one of these caves I can tell you it's breathtaking. A billion clanker slop images later and you can't produce ONE (fucking 1) of value. And you think I'm the one who's "lacking perspective"?

But once we have AGI and ASI the Minds will produce great art! Super related to current transformer architecture slop.

Was "Self‐Portrait as a Drowned Man" thought of as great art at the time it was taken in 1840? Or was its status as great art a retroactive judgement with the perspective of time by people after 1840?

Not being familiar with the piece, so I looked it up. Seems like it was a protest correspondence aimed at a specific institution, not a public artwork. It doesn't appear to have been exhibited, reviewed, or discussed in print in 1840. Mostly circulated privately.

So I don't think it's a good test. Because you're asking me to point out an equivalent piece in the billions of generative art produced in these six years that would be judged by future historians to have great artistic value. If we were both back in 1840, we both wouldn't be able to point out "Self‐Portrait as a Drowned Man" either. It never posed itself as art at the time, and we'd have to know what future art historians thought.

Yes, I stand by the "lacking perspective" comment, because your judgements on photography and what's considered art are from the current modern point of view, rather than putting yourself in the shoes of those judging photography when it first came out--which is the shoes we find ourselves wearing when it comes to judging generative images, because the alternative (retrospectively judging it from the future) is unavailable to us.

Hence, my core argument still stands. In what way is photography different than generated images, where in photography is considered art because humans can exercise their creative judgement, but in generative images isn't because humans cannot exercise their creative judgement?

To argue against my own stance, the tact you should have taken could have been:

If one major component of art is the human judgement, how could humans could be credited for exercising any judgement at all when prompting, when LLMs are at the end of the pipe? And if they're at all intelligence, what's to say they're not superseding our intentions with their own?

People hate the false effort signaling. Photos require effort and skill to take, and while it's perhaps less than an oil painting, it's still something.

If you push AI-generated content to make stuff look fancy, it's a bad idea. If you generate stuff to communicate more clearly, I don't see a problem.

Barely longer than a Tweet. Low effort slop. /half sarcasm.

I record 20-40 minutes of "uneditied, one take no fake" Podcast/Vlog per day and have been using AI to generate the cover image + video description and sometimes title for YouTube lately. Also somewhat low effort, I admit. I have project ideas to upcycle the raw takes. (iSpooge Daily on YouTube, support links on channel page).

My likes and views are consistently up since using Gen AI for said tasks. It's extremely niche, just yapping about existence and ideas as a homeless unemployed software engineer in Christ on daily walks. Some of the most authentic stuff online, in my opinion, criminally underrated. The AI art covers and generated metadata are better than I could do even given unlimited time + resources.

I think there's a difference here between low-effort AI slop and something made with the help of AI. I always wonder how many AI images and articles we consume without realizing they were made by AI just because a human actually spent the time on making the AI create a good output.

This is more my problem with it. Obvious AI usage tends to be slop content, so any sign of it becomes a useful heuristic as to whether the subject is useless spam/slop. But I'm not anti-AI. It's a category of tools that can be useful. I think using AI as part of the editing process can be very useful. Unfortunately it also enabled more spam. And if it looks like spam, I'll instinctively treat it as such.

Yeah exactly, I think the massive volume increase in content requires more aggressively filtering what you pay attention to. My tolerance for bad content has gone to nothing because the second I realize I can prompt the same article into existence except more personalized, I’m out - and the next time I see that company/person’s blog, I’m ignoring it.

Not everything made with some kind of AI assistance is slop, but all slop is made by AI.

563 upvotes for a 2 paragraph post of someone complaining about AI images in someone else's blogs?

This is "Bicycle face" level of hatred (or fear) against new technology.

I find images are good for speed-reading; they convey information that otherwise would require much more time-consuming attention. I'm significantly more likely to read a long article, after skimming it, if it contains contextual images.

Personally, I'm okay with AI generated images as long as it's not obvious or slop: For me, a carefully-prompted image is on par with a carefully chosen photo, carefully chosen diagram, ect.

I also hold blog entries to a lower standard than professional content: I'm much more likely (after skimming) to read an imperfect non-professional blog entry than (after skimming) imperfect professional content. (And professional content is where I expect images that are higher quality than AI prompting.)

I know it is overused on tweeter - but all this really seems to be about status. Like Robin Hanson said - people read stuff because it seems prestigious.

I agree. The human experience is the only thing that matters. If AI tries to replace or supersede that, then it's bad AI.

Agree. Ai art looks like garbage.

This might be true for technical users, average users increasingly don't care about AI images and even AI prose.

I stopped reading blog posts/tweets that have been AI generated.

OK.

So people should stop expressing themselves, however they see fit so you can read their blog? Ok!

AI images are the antithesis of self expression. Self expression would involve some actual creation.

I don't pick a book off a shelf and call it my own self expression. That's someone else's output. I also don't consider a machine's randomly generated output my own self expression. No actual person without financial incentive would.

You are missing the point, let's take "2001: A Space Odyssey" the novel book cover was not made by the writer Arthur C Clark. But by artist Robert McCall. Well not everyone can afford to hire an artist. You can be creative in your field, and still hire a human or use an AI to do art instead. What's the difference!

Using AI to generate content isn't expressing yourself. It's expressing non-human AI content that's actually not a part of your self - AI content is the opposite of self-expression, it's other-expression. That's why people dislike seeing it peddled as self-expression.

In traditional publishing publishers view the cover as a marketing tool - Arthur C Clark had negligible influence over what appeared on the covers of his books in his career when 2001 was released. That's not self-expression by any definition. You could hire a human to do art for you if you like, that's also not self-expression, nor is using an AI to do art for you. There's no meaningful self-expression in the act of having other people make things for you, they're doing the expression, you're enjoying (or not enjoying) the result.

Same thing with dark mode.

Weirdly I might slightly prefer your AI-generated image if some thought is put into it to entirely average soulless and tasteless corporate stock photo. You know multicultural team or photogenic enough person doing some meaningless staged task. Those have no chance to adding anything to the text. Same really goes for clip art.

At least with AI there might be some thought or idea behind it. Even it is lazy slop.

No one is reading your blog anyway, they are looking for the video.

If I do it, I make sure it's a good image, not just slop that I prompt and paste in, because I can. And on rare occasions when I do, it's on my own gaming system with Stable Diffusion so I can have as many takes as I like, fairly quickly.

Never do it because you can - that is a no-effort thing.

> I'd rather see a shitty Microsoft Paint drawing as opposed to some AI image.

So spot on!

And, with that Paint drawing, it'll probably illustrate what you mean and at the same time we establish some kind of trust; I'll know this is not just mere AI slop.

I for one, wish they don’t change what they’re doing, and that people don’t give them advice—let them use AI to generate anything that want.

That will help me easily distinguish and avoid bad bloggers (or any kind of merchants).

If I don't recognize that it's AI and it helps me understand the article I am fine with it. Generative AI images don't need to be slop even if most of them are.

I just assume that if you think it's fine to shit in my eyes with an AI image then you're probably also about to shit in my eyes a lot more with AI text.

you are but one person

I'm another one. There are at least two of us. I'd bet there's more.

Old man yell at cloud. I suppose it matters as long as old man cohort with purchasing power.

niceeee

Fuck it, I love em

Correction: Whenever you can tell it's an AI image, it discourages you

What I read , my western ego is so big I want to place my opinion everywhere possible .

Image is content , ai can generate art level images and an obvious slop . To read or or not is absolutely up to reader , not someone’s entitled opinion .

Isn't the point of a blog to "place my opinion everywhere possible"? Isn't that why there's an internet?

Still better than AI generated text.

You can't cater to every single person out there. The author comes as quite entitled.

what percentage of readers does it bother though? If 2% of readers just refuse anything built/generated with AI then are they worth accommodating? It sounds crazy but i distinctly remember people slamming their fist on the desk saying they will always prefer to use the card catalog at the library instead of Google when search was really taking off. At some point you just have to move on.

I share the author's sentiment.

But I have another hatred:

    Monospace fonts in blogs.

Monospace makes text harder to read and ugly. It was a necessity. It no longer is. Get over it.

If a content is making a positive impact on your life, your mental health, your family's well being, your general outlook, your future, would it matter if is from AI or from human?

> If it is making a positive impact, would it matter if it's AI or human

We're drowning in this stuff, and no one has enough hours in their day to find out if obviously AI-generated slop can make a positive impact on their life. Why don't you read it and let us know? We'll continue to not engage with it.

It doesn't seem to be though, and there's no reason to assume it will magically become so. We don't do the same with every other negative: suppose global warming made you immortal, wouldn't that be great.

Hmm... where have I encountered this worldview before

https://en.wikipedia.org/wiki/Degenerate_art

You heard it here first, not liking slop means you're a Nazi.

You heard it here for the thousandth time in the history of art: condescension towards new mediums does not make you morally superior, just conservative in a predictable and recurring way.

It's not a new medium, just a faster way of pumping out (mostly crappy versions of) the same thing.

Reply on news.ycombinator.com