Brilliant, per usual. What a wild reveal at the end!
I wrote something similar: “There is no test for AI contamination. There’s no blood sample for whether the sentence you just wrote was yours or whether your favorite author has now become a purveyor of slop or whether you’re unconsciously reproducing a pattern you absorbed from a thousand ChatGPT outputs.
The asterisk need not be applied to your work. It’s something that now exists in the atmosphere. You breathe it in whether you use the tools or not. And, like radiation, repeated exposure slowly kills you.”
The "cooling water leaking off GPU racks" line is genuinely one of the more striking images I've read lately. The question of intent is real — but what's interesting is that the Jevons dynamic applies here too: as the cost of generating text drops to zero, the attention premium on text that clearly required human thought goes up, not down. AI;DR might be less a boycott and more a new attention market forming in real time.
Yeah, but do GPU racks actually leak cooling water? I doubt it. Did you take it as a metaphor? You won't know by reading AI generated writing because they know even less. (They 'know' nothing.)
Haha god damn it Alberto 😂😂 point proven and well done!
What's surprised me is that social platforms don't have user-driven content labelling features to indicate if something was created by AI or not (with some level of differentiation on on the spectrum of what that actually means). The self-reporting isn't perfect, but it would be nice to know if something was AI going into it...
That was my stance, until I read this. I did derive insight from the post, but it was AI. If you labelled as AI-written, I would probably have skipped it.
Here's a question for you: When you do these AI written article experiments, how much effort would you say you apply to prompting and revising in a sort of 'creative director,' manner? Say that Substack had labels and there was an option to indicate the % human vs AI involvement, what would you label this (let's say, 10% increments)? The point I'm trying to get to is: at what rate of human involvement does writing feel human? Do you have a sense for it?
Well, I would say this is 90% me still haha. There's not that much AI writing. But there's some. That's almost always as it is with these experiments. AI can't write something like this no matter how much prompt care you put into it. The thing is that the 10% is completely invisible. Then it will be 20% then 30%. Most AI writing is instantly recognizable. Some will not be.
As for my personal view: I think around 20-30% I would still count it as human. As "acceptable". Ofc it depends which parts are outsourced (it's the argument, some sentences here and there, the main points, the headline, and so on) but if my AI-meter starts to fill up then that's a bad sign for me.
Thanks for the honesty, Alberto! But hey, 90% you - then the irony isn’t maxxed out at all, I would say.. then you tricked me twice!! 😄
And then, I totally agree: it matters A LOT which part is outsourced?
You write:
“Ofc it depends which parts are outsourced (it's the argument, some sentences here and there, the main points, the headline, and so on)”
If AI made the argument and the main points? Then how does that count in the percentage? I would say more than 10 percent - one could argue 60 or 70 percent.
Yeah. Certainty if you let AI make the main points then you better put a lot of "you" into the execution. However, if you outline everything and let AI execute for you, there's no way to accept that haha. AI can destroy the best of outlines.
Thank you for the reply. That makes sense to me; I write fiction and my rule has always been that AI doesn't get to put any words on the page. It's been a great research tool and thought partner, but I find that there's something about the creation process that I feel changed for in doing it. My thoughts have been that to the extent AI optimizes for that feeling is the extent to which it's use is worthwhile for that creator. Though, I don't let it into my writing, so I'm not sure what that says about me lol
Increasingly I take turgid human-written pieces and just get AI to summarize their key arguments -- which should not, of course, be necessary now that journalists no longer have to fill print columns.
Ive co written books 3 times (out of 9 books) AI is a better co author. Personally I dgaf about style unless it is fiction. Structured arguments. Close proximity to truth.
I loved this term. It hits very close to home. A useful practice I've found in today's world of AI slop is asking "is this worth 20 minutes of my attention?" before clicking anything. It's not about whether something is AI generated to me. There are several writers whose ideas are worth the time even when they are using AI. It's more about consistently evaluating the opportunity cost of reading something a machine might have generated in 45 seconds. Hard to know in advance, but worth a moment of reflection.
And not trying to be clever (truly), but when this debate crosses my inbox, feed, or conversation, I'm often struck by the question: is art, art because a human created it, or because a human was moved by it?
Interesting. I came back to comment here only after reading a later post where you mentioned there was a twist at the end of this article. I usually read and like all your articles but this one I only made it halfway through before exiting when it was first posted. I thought that makes this an interesting natural experiment for me as I should try and examine WHY I didn’t finish the article.
I didn’t exit because I thought it was AI.
I think in general I noted this article was less tight and more meandering than most of your others. I see that you indicated that 90% of content was you and only 10% of snippets are AI. Does that include the overall structure / outline / flow or did AI participate in that part?
I didn’t actually care about the topic and that contributed to my exiting midway. I don’t subscribe to AI;DR as a general principle (if content is still good I am OK with authors using tools). I thought the dialogue with Socrates was hilarious fun and interesting at the same time.
Yeah, I consider the Socrates one superior to this one in every way, yes; especially more tight as you say. (In my defense, this is an older draft, the other was very new!)
I love this! The comments have become the real sequel to the piece. They show how quickly people stop judging the work itself and start judging provenance, contamination, disclosure, percentages, legitimacy — all the moral theater around the artifact. That part of the experiment clearly worked. Where Romero weakens his own strongest position is by arguing too hard about whether it was 90 percent him, or that critics are proving his point, or that their reaction is mainly post-reveal bias. That shifts the center of gravity away from the strongest question and into a weaker fight about process and motive. The stronger line is the original one: here is the work. Does it live or not? Once the comments move off that ground, they mostly convict themselves. No need to help them change the subject.
> To a degree you’d rather not know, I am AI. And you did read this.
Would've bet a tenner on that, if I could have but found a sucker. Not out of anything in the tone or writing style or whatever, but because... c'mon, it's a trope verging on cliché for writing about AI writing.
Hahaha then you'd lose your bet! This is 90% written by me. No model can write something like this however hard you try. But there's some AI sprinkled here and there. Just invisible. (Btw, I invented that "cliche" so I'm allowed to exploit it!)
Ach, fair enough. (And I was trying to remember where I'd first seen it!)
> No model can write something like this however hard you try.
I do think it could be arranged, in principle, potentially with a hyper-specific mono-purpose model that would sacrifice aptitude at any other task. But the amount of effort you'd need to input into achieving this result would be "why the fuck even bother" levels. I certainly have neither the time, nor expertise, nor petty spite that would be needed for such an endeavor.
EDIT: The furries could do it, of course, if incentivized, but I can't imagine why they'd pursue such a task.
I just don’t want to read AI text. I’m not worried about the environmental impact. I don’t think it’s all hype. I just think it’s shit and I don’t want to read it. Is that allowed? And I got bored with this article halfway through so was not at all surprised to scroll down and see that you used AI to write it (wow, what a brilliant, Andy Kaufman-esque twist! Never done before!) In general your writing has gotten much worse since you’ve (obviously) started relying heavily on AI.
I can tell you didn't read it because this is just not true. You are proving my point. Actually a few of you are. Thank you. You got bored because writing that's not immediately easy makes you bored, probably. This isn't AI, I'm sorry I tricked you. There's only scattered traces that matter little. 90% of it is me. The fact that you assume it's all AI is very interesting. Worth reflecting upon
LOL, “you’re just not ready for my level of thinking”. Whatever bro. You wrote a boring, rambling piece I disagree with. Having skimmed the rest of it, it’s still boring and rambling and I still disagree with it. AI writing is never invisible, it’s never good, and anyone who recognises it is justified in deciding it’s not worth his time. And I say this as someone who uses AI heavily in professional contexts, including for writing.
I am arguing with your piece on its supposed merits, not on whether or not you used AI to write it. You chose to cling to that because it “proves your point”.
I can tell you didn't read it. Because there's a critical point he made..garbage in...garbage out. The output when writing with AI is a reflection of the input. If the output is trash, that's a reflection of the input. When writing with AI is done well, how is it any different from a good editor?
It’s different from a good editor in that it is a terrible editor, which takes good and original ideas and turns them into unreadable sludge like this article. This is the correct take, by someone who is not literally trying to sell you prompts https://fitzfromdublin.substack.com/p/you-are-the-slop?r=kuf0h&utm_medium=ios
Sharp point. We are moving from an information economy to a provenance economy. Soon readers will care less about whether text is correct than whether someone genuinely thought it. When content becomes infinite, sincerity becomes scarce. 
Brilliant, per usual. What a wild reveal at the end!
I wrote something similar: “There is no test for AI contamination. There’s no blood sample for whether the sentence you just wrote was yours or whether your favorite author has now become a purveyor of slop or whether you’re unconsciously reproducing a pattern you absorbed from a thousand ChatGPT outputs.
The asterisk need not be applied to your work. It’s something that now exists in the atmosphere. You breathe it in whether you use the tools or not. And, like radiation, repeated exposure slowly kills you.”
More: https://www.whitenoise.email/p/the-ai-asterisk
Thank you Tom. So good and an appropriate epistemic stance. The only one probably.
The "cooling water leaking off GPU racks" line is genuinely one of the more striking images I've read lately. The question of intent is real — but what's interesting is that the Jevons dynamic applies here too: as the cost of generating text drops to zero, the attention premium on text that clearly required human thought goes up, not down. AI;DR might be less a boycott and more a new attention market forming in real time.
Yeah, but do GPU racks actually leak cooling water? I doubt it. Did you take it as a metaphor? You won't know by reading AI generated writing because they know even less. (They 'know' nothing.)
Haha god damn it Alberto 😂😂 point proven and well done!
What's surprised me is that social platforms don't have user-driven content labelling features to indicate if something was created by AI or not (with some level of differentiation on on the spectrum of what that actually means). The self-reporting isn't perfect, but it would be nice to know if something was AI going into it...
That was my stance, until I read this. I did derive insight from the post, but it was AI. If you labelled as AI-written, I would probably have skipped it.
Here's a question for you: When you do these AI written article experiments, how much effort would you say you apply to prompting and revising in a sort of 'creative director,' manner? Say that Substack had labels and there was an option to indicate the % human vs AI involvement, what would you label this (let's say, 10% increments)? The point I'm trying to get to is: at what rate of human involvement does writing feel human? Do you have a sense for it?
Well, I would say this is 90% me still haha. There's not that much AI writing. But there's some. That's almost always as it is with these experiments. AI can't write something like this no matter how much prompt care you put into it. The thing is that the 10% is completely invisible. Then it will be 20% then 30%. Most AI writing is instantly recognizable. Some will not be.
As for my personal view: I think around 20-30% I would still count it as human. As "acceptable". Ofc it depends which parts are outsourced (it's the argument, some sentences here and there, the main points, the headline, and so on) but if my AI-meter starts to fill up then that's a bad sign for me.
Don't be a troll. :p
Thanks for the honesty, Alberto! But hey, 90% you - then the irony isn’t maxxed out at all, I would say.. then you tricked me twice!! 😄
And then, I totally agree: it matters A LOT which part is outsourced?
You write:
“Ofc it depends which parts are outsourced (it's the argument, some sentences here and there, the main points, the headline, and so on)”
If AI made the argument and the main points? Then how does that count in the percentage? I would say more than 10 percent - one could argue 60 or 70 percent.
Yeah. Certainty if you let AI make the main points then you better put a lot of "you" into the execution. However, if you outline everything and let AI execute for you, there's no way to accept that haha. AI can destroy the best of outlines.
Thank you for the reply. That makes sense to me; I write fiction and my rule has always been that AI doesn't get to put any words on the page. It's been a great research tool and thought partner, but I find that there's something about the creation process that I feel changed for in doing it. My thoughts have been that to the extent AI optimizes for that feeling is the extent to which it's use is worthwhile for that creator. Though, I don't let it into my writing, so I'm not sure what that says about me lol
Increasingly I take turgid human-written pieces and just get AI to summarize their key arguments -- which should not, of course, be necessary now that journalists no longer have to fill print columns.
Ive co written books 3 times (out of 9 books) AI is a better co author. Personally I dgaf about style unless it is fiction. Structured arguments. Close proximity to truth.
I loved this term. It hits very close to home. A useful practice I've found in today's world of AI slop is asking "is this worth 20 minutes of my attention?" before clicking anything. It's not about whether something is AI generated to me. There are several writers whose ideas are worth the time even when they are using AI. It's more about consistently evaluating the opportunity cost of reading something a machine might have generated in 45 seconds. Hard to know in advance, but worth a moment of reflection.
Bravo.
And not trying to be clever (truly), but when this debate crosses my inbox, feed, or conversation, I'm often struck by the question: is art, art because a human created it, or because a human was moved by it?
Interesting. I came back to comment here only after reading a later post where you mentioned there was a twist at the end of this article. I usually read and like all your articles but this one I only made it halfway through before exiting when it was first posted. I thought that makes this an interesting natural experiment for me as I should try and examine WHY I didn’t finish the article.
I didn’t exit because I thought it was AI.
I think in general I noted this article was less tight and more meandering than most of your others. I see that you indicated that 90% of content was you and only 10% of snippets are AI. Does that include the overall structure / outline / flow or did AI participate in that part?
I didn’t actually care about the topic and that contributed to my exiting midway. I don’t subscribe to AI;DR as a general principle (if content is still good I am OK with authors using tools). I thought the dialogue with Socrates was hilarious fun and interesting at the same time.
Yeah, I consider the Socrates one superior to this one in every way, yes; especially more tight as you say. (In my defense, this is an older draft, the other was very new!)
I love this! The comments have become the real sequel to the piece. They show how quickly people stop judging the work itself and start judging provenance, contamination, disclosure, percentages, legitimacy — all the moral theater around the artifact. That part of the experiment clearly worked. Where Romero weakens his own strongest position is by arguing too hard about whether it was 90 percent him, or that critics are proving his point, or that their reaction is mainly post-reveal bias. That shifts the center of gravity away from the strongest question and into a weaker fight about process and motive. The stronger line is the original one: here is the work. Does it live or not? Once the comments move off that ground, they mostly convict themselves. No need to help them change the subject.
That's the businessman in me in conflict with the artist haha
> To a degree you’d rather not know, I am AI. And you did read this.
Would've bet a tenner on that, if I could have but found a sucker. Not out of anything in the tone or writing style or whatever, but because... c'mon, it's a trope verging on cliché for writing about AI writing.
Hahaha then you'd lose your bet! This is 90% written by me. No model can write something like this however hard you try. But there's some AI sprinkled here and there. Just invisible. (Btw, I invented that "cliche" so I'm allowed to exploit it!)
Ach, fair enough. (And I was trying to remember where I'd first seen it!)
> No model can write something like this however hard you try.
I do think it could be arranged, in principle, potentially with a hyper-specific mono-purpose model that would sacrifice aptitude at any other task. But the amount of effort you'd need to input into achieving this result would be "why the fuck even bother" levels. I certainly have neither the time, nor expertise, nor petty spite that would be needed for such an endeavor.
EDIT: The furries could do it, of course, if incentivized, but I can't imagine why they'd pursue such a task.
I genuinely don't think this is possible today. I invite anyone to try. It's not a matter of anything. The models can't do this.
Did you write this with AI? I tried to read the whole thing but started skimming half way through….
No, why? Did you catch any sentence you thought was AI? Please, let me know which!
Joke's on you, I *didn't* read.
Too scattered and surface-level, didn't seem to have a clear point, so I skipped to the end and lo and behold!
Slippity Sloppity
It wouldn't be great if the comments were also AI? ;-)
Half are hahaha
I just don’t want to read AI text. I’m not worried about the environmental impact. I don’t think it’s all hype. I just think it’s shit and I don’t want to read it. Is that allowed? And I got bored with this article halfway through so was not at all surprised to scroll down and see that you used AI to write it (wow, what a brilliant, Andy Kaufman-esque twist! Never done before!) In general your writing has gotten much worse since you’ve (obviously) started relying heavily on AI.
I can tell you didn't read it because this is just not true. You are proving my point. Actually a few of you are. Thank you. You got bored because writing that's not immediately easy makes you bored, probably. This isn't AI, I'm sorry I tricked you. There's only scattered traces that matter little. 90% of it is me. The fact that you assume it's all AI is very interesting. Worth reflecting upon
If it's not AI, and I can believe it, were you on a harsh deadline? Because it was not up to your usual standard.
LOL, “you’re just not ready for my level of thinking”. Whatever bro. You wrote a boring, rambling piece I disagree with. Having skimmed the rest of it, it’s still boring and rambling and I still disagree with it. AI writing is never invisible, it’s never good, and anyone who recognises it is justified in deciding it’s not worth his time. And I say this as someone who uses AI heavily in professional contexts, including for writing.
I'm being truthful. You can do with that what you wish
I am arguing with your piece on its supposed merits, not on whether or not you used AI to write it. You chose to cling to that because it “proves your point”.
To be clear I never assumed or claimed “it’s all AI”. But the traces are there and they make your writing worse, and not just in this specific piece
I can tell you didn't read it. Because there's a critical point he made..garbage in...garbage out. The output when writing with AI is a reflection of the input. If the output is trash, that's a reflection of the input. When writing with AI is done well, how is it any different from a good editor?
Good point. I don't think we know yet if AI can be a good editor; I do think AI makes a weird writer. I always get an uncanny valley chill from it.
It’s different from a good editor in that it is a terrible editor, which takes good and original ideas and turns them into unreadable sludge like this article. This is the correct take, by someone who is not literally trying to sell you prompts https://fitzfromdublin.substack.com/p/you-are-the-slop?r=kuf0h&utm_medium=ios
Er. Because good editors know what they are about and AI doesn't know anything at all because it is not cogent? Just a guess.
Sharp point. We are moving from an information economy to a provenance economy. Soon readers will care less about whether text is correct than whether someone genuinely thought it. When content becomes infinite, sincerity becomes scarce. 
The semicolon switch is the tell. TL;DR was reader laziness. AI;DR is reader distrust. Different problem entirely.