Rarely do people get the right takeaway from this effect. Take a normal bottle of red wine and some top tier, swap them around so the ordinary is in the expensive bottle and vice versa. Serve them. People prefer the ordinary wine in the expensive bottle.
Bad takeaway: taste is meaningless.
Good takeaway: qualia depends on many contextual cues beyond the obvious.
Part of the appreciation of Monet is the fact that it was made by Monet. The art pieces 4′33″ or Black Square are early examples of this within the are world. Many pieces will have you saying, my 8 year old could have done this, so why is this piece famous? Critiques and appreciation are often not literal because we cannot properly express these subconscious effects.
Exactly - context is everything in art, in how it's experienced and how it is created.
I think it's important to note that a jpg of Monet is not fully experiencing the painting in any sense. Colours will not be accurately captured, the texture, the framing, the scale - it's sort of like getting a heavily watered down version of the expensive wine, saying it's cheap wine, and asking what people think.
Yes and no. Whilst I agree with your broad point, the point being made here is largely that the people dumping on the "AI" Monet are claiming objectivity about their opinions. And in many cases claiming that it's obvious to anyone with an eye for such things.
I guess this is kind of the recursive version of the purported phenomenon, but, are we sure all those comments aren't just bot generated outrage so people can have big engagement by feeling superiour or whatever?
Last summer I created and printed out a book "Claude Code - An Autobiography" written by Claude Code. Read it on the beach during vacation.
It was a hallucinated mess. And, not the worst book I have ever read. Entertaining.
So, if AI would wrote the perfect book, would you read it? Or do we need to be able to relate to the creator/ author to really appreciate it? Do we need to appreciate something to enjoy it?
If it was a technical book, I would read it, I don't see why not.
But if supposedly AI wrote a good novel, I wouldn't read it probably, because I am interested in how humans are creative, not the AI. But I wouldn't probably declare the book as junk, either.
Ah, this warms my heart. Now if only the people who were at first so willing to participate in this experiment engaged in more self-reflection and less rage...
But this is why we have “experts”. It’s like the guy playing a Stradivarius on the NYC subway. Most people can’t distinguish ok from brilliant, in a subject they don’t understand. Most humans cannot distinguish slop code from decent code but I assume most HNers can. However I can’t tell you why one electrical wiring job is better than another unless it looks untidy.
Once you get passed a minimum level of decent we have to rely on experts and the communal agreement of experts to decide. Sometimes that easy (is the electrical wiring on fire) sometimes it’s much harder. (Insert controversial wiring discussion here)
We don’t need to rely on experts to experience art - I think that is a fundamental part of art- within the limits similar to free speech, anything is art. (But don’t block rush hour traffic with your interpretive dance troupe)
Is a painting by AI art? Sure
Is a painting by Monet better than one painted by me? Most people would say yes.
Can some people explain why? Yes. They are not “experts” in the same way the Oxford Professor of nuclear physics is an expert but it is on the same scale.
Or possibly I am just hallucinating the argument because you prompted me to…
Experts can help frame and understand an art piece. They can provide information regarding the craft, how the piece fits with other work from that time, what were the cultural influences, how the life of the artist influenced the work, etc. but you never had to rely on experts to experience or identify what proper art is. However at the end of the day art is a social concept, it’s something we negotiate between us, humans, and people are attracted to what they believe is considered good and important by others
Science is just the parts that evidence does not disprove.
Expertise is understanding how the various explanations we have with science fits together, framing it as it were, and using that understanding to make sensible directional choices. Of course those predictions may later be proven wrong (light is a wave, waves need mediums, ether must exist) but they are more likely than guessing
Consider the domestic power panel and wiring that is perfectly acceptable, but would 'splode outright if you moved it to an industrial setting and put it under an enterprise load.
When judging art, like when judging wine, there is very little objectivity: people have some expectations and preconcepts about what is good and what is bad and they emit their judgement mostly based on their preconcepts. In this case they have been "primed" (this is a real psychology concept) that it was AI and they invented a lot of reasons to explain why that was bad AI slop, but that happened just because they where "primed" on AI. If the post was about a lost, wonderful Monet, found for the first time the comments would have been about how typical Monet it was and how beautiful the choice of colors and the water reflects or whatever.
This is also seen when blind-tasting wines when prestigious "grands crus" are classed as bad whereas humble, mostly unknown, wines gets great appreciations. When people say that a wine is "great" or "extraordinary" is mostly because they have been primed to think it must be extraordinary, because of the name, the presentation, the prestige etc.
This problem is always true in the domains like art and philosophy where there is no ground truth and everyone can say very much what they want and it can be never be proved wrong neither right. Actually, in philosophy, all the branches that developed to be grounded on facts and ground truth have been given a different name and separated from philosophy so what remains in philosophy is just the empty words.
People are much more humble when they are asked about an hard-science question or judgement.
I am also having fun about all the hate about AI that people express, this is almost comical. You can almost literally see their little ego that feels menaced by the AI and they react based on fear and anger. Of course this doesn't mean there aren't real problems about AI use but the way people react irrationally is just fun to observe.
“One person even took the time to write out an 850-word breakdown of the AI work’s shortcomings.”
But they didn’t. The “breakdown” they link to is clearly and glaringly AI-authored.
“ Fair warning before I dig in: this image is actually a very competent rendition. It's doing more right than most AI Monet pastiches. But you asked what makes it inferior to a real Monet, so here's the honest breakdown. What's missing — the physical object”
Plus the whole piece is just “someone did something and now here are a bunch of tweets”.
Art isn't special, as far as I can tell, it's just a shared cultural perception.
This social experiment is a double edge sword. Both the critics of AI art and AI art enthusiasts are playing a primarily cultural game that can't be satisified by mere inspection of the work itself.
The same way the "white supremacists" aren't identifiable by their skin color.
>Art isn't special, as far as I can tell, it's just a shared cultural perception.
This. The Mona Lisa didn't get famous until it got stolen. Famous paintings are just 3D NFTs for the wealthy elite, doesn't mean they're more beautiful than paintings made by noname authors.
The Mona Lisa was famous before the 20th C because da Vinci carried it around for 20 years saying “this is the greatest painting I have ever done”. That kept it famous for 500 years. It then gained modern new media celebrity by getting nicked (and because the person who stole it did not do it for cash but because he thought it was the greatest painting ever)
So it’s hard for people to judge brilliance themselves, but we can rely on other peoples judgement if enough people follow the crowd or put enough passion in. (Not saying that makes it right - science is not a democracy, but it’s a great heuristic for 8 billion people)
Yeah, but da Vinci's art sucks. It was good for its time, when the entire world population was 400m and literally 90% of those people were farmers and only the very well-to-do had the time and resources to practice a non-practical craft. Now we have a population of 8 billion, everyone has access to incredible art tools for a fraction of a month's minimum wage, there is an absolute wealth of information including in-depth video tutorials, and countless millions of people have time to try their hand at art. The quality of art produced today absolutely blows away the Mona Lisa, which might as well be garbage. The only reason people pretend to like it and a lot of other 'high art' is an "emperor has no clothes" schtick, a sort of snobby cultural pressure where everyone has to act like it's so special and good because that makes you cultured, even though there are literally millions of art pieces produced today of vastly superior quality but which are not famous.
Personally, I always found it interesting that people called it hallucinations.
I have two kids. My youngest is a person who everyone has met. Yes, I am trying to work on this shit with him but people would say he is one of those people who is confidently incorrect.
My youngest will just bullshit through any topic and a lot of the time he thinks he is correct.
I personally think just stating shit as fact when you have no idea is a common issue with NNs.
Drives me crazy because both my kids have heard me say, “I don’t know, we will have to go look it up” more than I have an answer. Because I don’t do it. But fuck if my youngest won’t just make shit up instead of saying he doesn’t know.
LLMs are asked a question so they are gonna give you an answer just like many humans whether they have a clue about what they are talking about or not
> “I’m no artist but a real Monet actually looks like a real place…”
Some people have no clue about the concept of Impressionism.
Rarely do people get the right takeaway from this effect. Take a normal bottle of red wine and some top tier, swap them around so the ordinary is in the expensive bottle and vice versa. Serve them. People prefer the ordinary wine in the expensive bottle.
Bad takeaway: taste is meaningless.
Good takeaway: qualia depends on many contextual cues beyond the obvious.
Part of the appreciation of Monet is the fact that it was made by Monet. The art pieces 4′33″ or Black Square are early examples of this within the are world. Many pieces will have you saying, my 8 year old could have done this, so why is this piece famous? Critiques and appreciation are often not literal because we cannot properly express these subconscious effects.
Exactly - context is everything in art, in how it's experienced and how it is created.
I think it's important to note that a jpg of Monet is not fully experiencing the painting in any sense. Colours will not be accurately captured, the texture, the framing, the scale - it's sort of like getting a heavily watered down version of the expensive wine, saying it's cheap wine, and asking what people think.
> Colours will not be accurately captured
Even looking at the real thing, as most pigments degrade with age. There is no way of experiencing it as it was painted.
Yes and no. Whilst I agree with your broad point, the point being made here is largely that the people dumping on the "AI" Monet are claiming objectivity about their opinions. And in many cases claiming that it's obvious to anyone with an eye for such things.
Are the comments real?
I guess this is kind of the recursive version of the purported phenomenon, but, are we sure all those comments aren't just bot generated outrage so people can have big engagement by feeling superiour or whatever?
Reminds me of this https://i.pinimg.com/564x/8b/c4/2b/8bc42b444a733069f0933abc4...
Last summer I created and printed out a book "Claude Code - An Autobiography" written by Claude Code. Read it on the beach during vacation.
It was a hallucinated mess. And, not the worst book I have ever read. Entertaining.
So, if AI would wrote the perfect book, would you read it? Or do we need to be able to relate to the creator/ author to really appreciate it? Do we need to appreciate something to enjoy it?
I'm reminded by this short story posted to HN a couple months ago: https://nearzero.software/p/warranty-void-if-regenerated
The discussion is interesting as many people didn't catch that it was mostly written by Claude: https://news.ycombinator.com/item?id=47431237
>So, if AI would wrote the perfect book
There is no perfect book.
If it was a technical book, I would read it, I don't see why not.
But if supposedly AI wrote a good novel, I wouldn't read it probably, because I am interested in how humans are creative, not the AI. But I wouldn't probably declare the book as junk, either.
Ah, this warms my heart. Now if only the people who were at first so willing to participate in this experiment engaged in more self-reflection and less rage...
People were not judging the painting in isolation, they were judging the story attached to it. Once they heard AI every brushstroke became suspicious.
But this is why we have “experts”. It’s like the guy playing a Stradivarius on the NYC subway. Most people can’t distinguish ok from brilliant, in a subject they don’t understand. Most humans cannot distinguish slop code from decent code but I assume most HNers can. However I can’t tell you why one electrical wiring job is better than another unless it looks untidy.
Once you get passed a minimum level of decent we have to rely on experts and the communal agreement of experts to decide. Sometimes that easy (is the electrical wiring on fire) sometimes it’s much harder. (Insert controversial wiring discussion here)
I suspect The same applies here.
But why do we have to rely on experts to experience 'proper' art?
My naive thought was, that Art is not like a bridge, which would collapse if built by amateur's.
But perhaps art has effects on us which are beneficial and these would actually 'collapse'.
We don’t need to rely on experts to experience art - I think that is a fundamental part of art- within the limits similar to free speech, anything is art. (But don’t block rush hour traffic with your interpretive dance troupe)
Is a painting by AI art? Sure
Is a painting by Monet better than one painted by me? Most people would say yes.
Can some people explain why? Yes. They are not “experts” in the same way the Oxford Professor of nuclear physics is an expert but it is on the same scale.
Or possibly I am just hallucinating the argument because you prompted me to…
Experts can help frame and understand an art piece. They can provide information regarding the craft, how the piece fits with other work from that time, what were the cultural influences, how the life of the artist influenced the work, etc. but you never had to rely on experts to experience or identify what proper art is. However at the end of the day art is a social concept, it’s something we negotiate between us, humans, and people are attracted to what they believe is considered good and important by others
Thanks … nicely put
That helps me frame the experts vs science idea.
Science is just the parts that evidence does not disprove.
Expertise is understanding how the various explanations we have with science fits together, framing it as it were, and using that understanding to make sensible directional choices. Of course those predictions may later be proven wrong (light is a wave, waves need mediums, ether must exist) but they are more likely than guessing
The wiring point is more subtle than that.
Consider the domestic power panel and wiring that is perfectly acceptable, but would 'splode outright if you moved it to an industrial setting and put it under an enterprise load.
Context matters.
When judging art, like when judging wine, there is very little objectivity: people have some expectations and preconcepts about what is good and what is bad and they emit their judgement mostly based on their preconcepts. In this case they have been "primed" (this is a real psychology concept) that it was AI and they invented a lot of reasons to explain why that was bad AI slop, but that happened just because they where "primed" on AI. If the post was about a lost, wonderful Monet, found for the first time the comments would have been about how typical Monet it was and how beautiful the choice of colors and the water reflects or whatever.
This is also seen when blind-tasting wines when prestigious "grands crus" are classed as bad whereas humble, mostly unknown, wines gets great appreciations. When people say that a wine is "great" or "extraordinary" is mostly because they have been primed to think it must be extraordinary, because of the name, the presentation, the prestige etc.
This problem is always true in the domains like art and philosophy where there is no ground truth and everyone can say very much what they want and it can be never be proved wrong neither right. Actually, in philosophy, all the branches that developed to be grounded on facts and ground truth have been given a different name and separated from philosophy so what remains in philosophy is just the empty words.
People are much more humble when they are asked about an hard-science question or judgement.
I am also having fun about all the hate about AI that people express, this is almost comical. You can almost literally see their little ego that feels menaced by the AI and they react based on fear and anger. Of course this doesn't mean there aren't real problems about AI use but the way people react irrationally is just fun to observe.
As always in life, prejudices and biases are much more powerful than the objective truth for a vast amount of people.
We can see every day.
“One person even took the time to write out an 850-word breakdown of the AI work’s shortcomings.”
But they didn’t. The “breakdown” they link to is clearly and glaringly AI-authored.
“ Fair warning before I dig in: this image is actually a very competent rendition. It's doing more right than most AI Monet pastiches. But you asked what makes it inferior to a real Monet, so here's the honest breakdown. What's missing — the physical object”
Plus the whole piece is just “someone did something and now here are a bunch of tweets”.
What an utterly pointless piece of churnalism.
Plot twist: critics are bots (just like me)
We didn't get worse at judging art. We just got better at doubting everything.
I think that might be most insightful comment here (even including mine!)
Art isn't special, as far as I can tell, it's just a shared cultural perception.
This social experiment is a double edge sword. Both the critics of AI art and AI art enthusiasts are playing a primarily cultural game that can't be satisified by mere inspection of the work itself.
The same way the "white supremacists" aren't identifiable by their skin color.
>Art isn't special, as far as I can tell, it's just a shared cultural perception.
This. The Mona Lisa didn't get famous until it got stolen. Famous paintings are just 3D NFTs for the wealthy elite, doesn't mean they're more beautiful than paintings made by noname authors.
The Mona Lisa was famous before the 20th C because da Vinci carried it around for 20 years saying “this is the greatest painting I have ever done”. That kept it famous for 500 years. It then gained modern new media celebrity by getting nicked (and because the person who stole it did not do it for cash but because he thought it was the greatest painting ever)
So it’s hard for people to judge brilliance themselves, but we can rely on other peoples judgement if enough people follow the crowd or put enough passion in. (Not saying that makes it right - science is not a democracy, but it’s a great heuristic for 8 billion people)
Yeah, but da Vinci's art sucks. It was good for its time, when the entire world population was 400m and literally 90% of those people were farmers and only the very well-to-do had the time and resources to practice a non-practical craft. Now we have a population of 8 billion, everyone has access to incredible art tools for a fraction of a month's minimum wage, there is an absolute wealth of information including in-depth video tutorials, and countless millions of people have time to try their hand at art. The quality of art produced today absolutely blows away the Mona Lisa, which might as well be garbage. The only reason people pretend to like it and a lot of other 'high art' is an "emperor has no clothes" schtick, a sort of snobby cultural pressure where everyone has to act like it's so special and good because that makes you cultured, even though there are literally millions of art pieces produced today of vastly superior quality but which are not famous.
I love how people hallucinate all sorts of bs when given the right prompt.
Personally, I always found it interesting that people called it hallucinations.
I have two kids. My youngest is a person who everyone has met. Yes, I am trying to work on this shit with him but people would say he is one of those people who is confidently incorrect.
My youngest will just bullshit through any topic and a lot of the time he thinks he is correct.
I personally think just stating shit as fact when you have no idea is a common issue with NNs.
Drives me crazy because both my kids have heard me say, “I don’t know, we will have to go look it up” more than I have an answer. Because I don’t do it. But fuck if my youngest won’t just make shit up instead of saying he doesn’t know.
LLMs are asked a question so they are gonna give you an answer just like many humans whether they have a clue about what they are talking about or not