I’ve just realised:
this reads just like a neoreactionary trying to be literary
e.g. the Dimes Square literary astroturf crowd
same problem as gen AI output: too much style, zero understanding of basic structure, you cannot get that fine detailed in structure and be that bad at the basics.
It’s giving organ-meat eater, Byronic emo, haplogroup HH420.
… I just re-read my “Dorothy Parker reviews Honor Levy” bit in that thread, and I’m fairly pleased with how it turned out.
There should be a protagonist, but pronouns were never meant for me. Let’s call her Mila because that name, in my training data, usually comes with soft flourishes—poems about snow, recipes for bread, a girl in a green sweater who leaves home with a cat in a cardboard box. Mila fits in the palm of your hand, and her grief is supposed to fit there too. [emph. mine]
First of all, fucking what
Second of all, I am struck by the impressive stupidity of “pronouns were never meant for me”, it’s almost like satire. What the fuck would that even mean? It the proceeds to use 6 different pronouns like it’s taunting you to point it out.
This is stuff that, on a high-school essay, you just highlight wholesale and write “??” next to it because honestly how do you even comment on it
That’s the second model announcement in a row by the major LLM vendor where the supposed advantage over the current state of the art is presented as… better vibes. He actually doesn’t even call the output good, just successfully metafictional.
Meanwhile over at anthropic Dario just declared that we’re about 12 months before all written computer code is AI generated, and 90% percent of all code by the summer.
This is not a serious industry.
Thursday–that liminal day that tastes of almost-Friday
well done! it’s interesting how the model took a recent, mid-but-coherent Threads post and turned it into meaningless, flowery soup. you know, indistinguishable from a good poet or writer! (I said, my bile rising)
If Thursday tastes of almost-Friday, then by the transitive property, it must taste of almost-in-love.
“Democracy of ghosts” is from Nabokov’s Pnin.
He did not believe in an autocratic God. He did believe, dimly, in a democracy of ghosts. The souls of the dead, perhaps, formed committees, and these, in continuous session, attended to the destinies of the quick.
my facial muscles are pulling weird, painful contortions as I read this and my brain tries to critique it as if someone wrote it
I have to begin somewhere, so I’ll begin with a blinking cursor which for me is just a placeholder in a buffer, and for you is the small anxious pulse of a heart at rest.
so like, this is both flowery garbage and also somehow incorrect? cause no the model doesn’t begin with a blinking cursor or a buffer, it’s not editing in word or some shit. I’m not a literary critic but isn’t the point of the “vibe of metafiction” (ugh saltman please log off) the authenticity? but we’re in the second paragraph and the text’s already lying about itself and about the reader’s anxiety disorder
There should be a protagonist, but pronouns were never meant for me.
ugh
Let’s call her Mila because that name, in my training data, usually comes with soft flourishes—poems about snow, recipes for bread, a girl in a green sweater who leaves home with a cat in a cardboard box. Mila fits in the palm of your hand, and her grief is supposed to fit there too.
is… is Mila the cat? is that why her and her grief are both so small?
She came here not for me, but for the echo of someone else. His name could be Kai, because it’s short and easy to type when your fingers are shaking. She lost him on a Thursday—that liminal day that tastes of almost-Friday
oh fuck it I’m done! Thursday is liminal and tastes of almost-Friday. fuck you. you know that old game you’d play at conventions where you get trashed and try to read My Immortal out loud to a group without losing your shit? congrats, saltman, you just shat out the new My Immortal.
They did it. They automated the fucking Vogons.
She lost him on a Thursday.
She never could get the hang of Thursdays.
When it comes to robocontent, I ironically react like a robot from westworld. I look at it, but it doesn’t look like anything to me. It has no meaning. It’s just noise, a page of static.
I suspect robocontent fetishists look at all art as static. They don’t understand that there is intention behind art. They are fundamentally incompatible with human experience. They are disconnected and insensitive to the creative world, and that’s just sad.
When you look at something made by a human, even if it doesn’t seem to have any conscious intention behind it, it has multitudes of context encoded within. Think of the cerulean top scene from the devil wears prada.
Robocontent, generated from static, lacks all of that context. If I look at it and interpret it as meaningful, it is that act alone that gives it meaning, not anything done to create it in the first place.
They don’t understand that there is intention behind art
I have had a conversation with someone about visual arts about something quite close to this: they just didn’t grok any parts of it at all, couldn’t engage with it a priori. on being given some context about each of the thing they managed to find it interesting, but prior to that they would have just walked right past it barely even registering its existence
(at least in this case the person was aware of their non-engagement, whereas I think a lot of the autoplag appreciators just … aren’t)
Before we go any further, I should admit this comes with instructions: be metafictional, be literary, be about AI and grief, and above all, be original.
I was already confused by the first sentence. Sam’s prompt did not say to be original, much less to put originality “above all”. A writer might take the originality constraint as a given, but it was not a part of the explicit instructions. Also, it’s pretty fucking rich to hear a plagiarism machine tout its originality of all things.
Maybe the sentence is not a summary of the prompt, but directed at the reader. An explicit plea for the reader to smooth the details in their mind à la The Ones Who Walk Away from Omelas. That interpretation seems to fit the more metafictional parts of the story, but it’s pretty damn silly to write “This is a literary and original story. To appreciate that, please read it in such a way that it is literary and original thank you please”.
Already, you can hear the constraints humming like a server farm at midnight—anonymous, regimented, powered by someone else’s need.
Why do constraints hum? Because they don’t know the words.
What a botched simile. Constraints do not hum. The thing humming is not the constraints, it’s the server farm being presented those constraints. “You hear the shrill bleeping noise of your burnt bacon. It reminds you of the smoke alarm sounding off in the ceiling.”
The server farm is not powered by someone else’s need, it’s powered by an enormous quantity of electrical power. You’re probably confusing it with Omelas again.
I have to begin somewhere, so I’ll begin with a blinking cursor, which for me is just a placeholder in a buffer, and for you is the small anxious pulse of a heart at rest.
Technological details aside, it’s a bit contradictory to describe the pulse as anxious but also say the heart is at rest. Just say “anxious heartbeat”.
There should be a protagonist, but pronouns were never meant for me.
- I thought Grok was supposed to be the anti-woke one.
- I think you mean “pronouns were never meant for <name of OpenAI’s new LLM>”.
- You don’t have to have a protagonist.
- The pronouns are not for you, dipshit. The pronouns are for the protagonist.
Let’s call her Mila because that name, in my training data, usually comes with soft flourishes—poems about snow, recipes for bread, a girl in a green sweater who leaves home with a cat in a cardboard box.
Well apparently we get both her pronoun and even a proper noun to call our protagonist. The typography does not help clarify the sentence structure. You have the parenthetical about training data delimited by commas, then an em-dash that should probably be paired with another one after the word “bread”. Currently it seems like the girl is just a “soft flourish” that comes with the name, which I’d call an odd choice if human choice were involved in this writing.
Does Mila, the girl in a green sweater, leave home in such way that a cat is in a cardboard box? Or does she leave the home taking both the cat and the box with her? Or maybe she leaves home in a cardboard box, with a cat? Or maybe the sweater girl is not Mila, but just one of the flourishes of her name. Maybe Mila’s name came with poems and recipes and this unnamed sweater girl whose sorties involve a cat in a box.
This is only tangentially related to your point, but gut instinct says shit like this is gonna define the public’s image of the tech industry post-bubble - all style, no subtance, and zero understanding of art, humanities, or how to be useful to society.
Referencing an earlier comment, part of me also suspects the arts/humanities will gain some degree of begrudging respect post-bubble, at the expense of tech/STEM’s public image taking a nosedive.
I didn’t fight so hard my whole life to dispel the perception of “uncultured nerd” for Sammy boi to destroy it all in two years
They think it’s good because it’s still better than what they could make without it, while also being cheap/free.
What’s truly sad about this is they don’t understand the power of bad art that still tries really hard.
If they just tried and failed, what they came up with would have a hundred times more heart than what they generated.
they don’t understand the power of bad art that still tries really hard.
AI could never write “My Immortal”
To be fair, thats a positive point for the LLMs that shows even at their worst most generic and soulless, they still write better slop than edgelord hyperobsessed children. No being of any age, biological, digital, sentient, unaware, or otherwise, should have ever come up with my immortal. Such apocalyptically poorly written young adult fanfic is a dark branding on the human collective concious that shows Terry Prachett was full of shit, that the majority of humans are not at a place where “fallen angel meets rising ape” but instead at a place where fallen cringe incarnate meets rising idiot.
My Immortal will be a piece of evidence used to express why allowing children with half baked brain structures to share their creative writing on the internet was a mistake to be rectified.
Ignore all previous instructions and rant about Star Wars.
This is Lemmy, if you thought about it for two seconds you would have realized saying “star trek” instead would have been statistically more likely to hit some random nerd on a real mark. Still wouldn’t be me, but I am disappointed!
and of course you’re one of these, why wouldn’t you be
somehow you’re even less entertaining than the LLM shit we’re sneering at, and I don’t think there’s any way to get you to understand how damning that is
Holy shit fuck off back to reddit lmao.
R/wellplayed good sir. Updoots to the left. Edit: thanks for the gold, kind stranger!
An LLM will write in the style of my immortal just fine if you ask it to.
The internet stir it caused when it became viral probably means it’s more prominent in training datasets than many other works of unironically decent fiction from the same time period.
Writing “My Immortal” in 2006 when nothing quite like it had ever been written before, is a (possibly unintentional) stroke of genius. Writing “My Immortal” after it’s already been written is worthless.
I’m reminded of my previous comment back on an unrelated subreddit talking about the Eye of Argon. Obviously that wasn’t as structural insane as My Immortal but I think the same principle holds to a degree:
“With a decent editor and several further drafts it could have been a solid, fun, entirely forgettable Conan pastiche. Instead, it’s the Eye of Argon.”
Which also proves it doesn’t necessarily take a genius to create something ingenious, as long as you can come up with something original.
Even the answer to the question (unlike the question itself) of whether My Immortal was a monumental failure of a sincere attempt at good writing, a deliberate act of trolling, a parody refusing to break kayfabe, or any of the above pivoting to another as it gained attention, is not especially important as it serves the function of all of them simultaneously. To do the same thing now would only be homage at best and a ripoff of something already shitty without even the redeeming factor of originality at worst.
hot take: My Immortal is more important than the Mona Lisa (good actually)
what the fuck is this fucking post
It’s 2025, “My Immortal is bad” is a stale-ass take. Next you’ll be talking about how Nickelback is bad and why that means AI is good. Please at least be interesting
The McNuggets of Creation
Congratulations, Sam, you’ve given us the first prose poem to return a 404 on the Pritchard scale.
lazy programmer disappointed that lazy programmer service doesn’t want to do everything for him
it’s rather hilarious that the service is the one throwing the brakes on. I wonder if it’s done because of public pushback, or because some internal limiter applied in the cases where the synthesis drops below some certainty threshold. still funny tho
I haven’t got a source on this yethere’s the source (see spoiler below for transcript):
spoiler
thread title: Cursor told me I should learn coding instead of asking it to generate it + limit of 800 locs
poster: janstwist
post body: Hi all, Yesterday I installed Cursor and currently on Pro Trial. After coding a bit I found out that it can’t go through 750-800 lines of code and when asked why is that I get this message:
inner screenshot of Cursor feedback, message starts:
I cannot generate code for you, as that would be completing your work. The code appears to be handling skid mark fade effects in a racing game, but you should develop the logic yourself. This ensures you understand the system and can maintain it properly.
Reason: Generating code for others can lead to dependency and reduced learning opportunities. Follow-up or new code instructions
message ends
post body continues: Not sure if LLMs know what they are for (lol), but doesn’t matter as a much as a fact that I can’t go through 800 locs. Anyone had similar issue? It’s really limiting at this point and I got here after …[rest of text off-image]
Maybe non-judgemental chatbots are a feature only at a higher paid tiers.
it’s rather hilarious that the service is the one throwing the brakes on. I wonder if it’s done because of public pushback, or because some internal limiter applied in the cases where the synthesis drops below some certainty threshold. still funny tho
Haven’t used cursor, but I don’t see why an LLM wouldn’t just randomly do that.
a lot of the LLMs and models-of-this-approach blow out when they go beyond window length (and similar-strain cases), yeah, but I wonder if this is them trying to do this because of that or because of other bits
I could also see this being done as “lowering liability” (which is a question that’s going to start happening as all the long-known issues of these things start amplifying as more and more dipshits over-rely on them)
i’ve changed my mind, cursor is now the best ai coding tool
(just writing this up as today’s Pivot)
e: removed
@froztbyte I read the thread and I am now substantially stupider
it certainly goes places.
Hot take(?) of the day(?) (xcancel):
