Three writers jump in a Google doc to discuss the fallout from this most recent literary scandal and debate the future of the industry in the shadow of AI
Quite a few comments here are hedging by pretending it isn't possible for an LLM to generate text without lots of human work. This is simply not the case. You can go to ChatGPT right now and ask it to "generate a prompt for a poem" and then use that prompt to generate the poem. If you believe that still makes the prompter an artist, well, I respect the intellectual consistency I suppose! But it is dodging the real question to pretend the only LLM art out there involves tons of human effort on behalf of the prompter.
A poem is one thing, a whole book is an entirely different beast. Go to ChatGPT now and ask it to write one chapter of a book, sit and laugh. It can’t. It will give you 5k words but 70% if not more will be one-liners.
LLMs cannot write books. They cannot write even one good chapter of a book. It doesn’t have the computational power to do so or the experience to weave through the intricacies of books. Please try it yourself and let us know
If you agree LLMs can already write poems and short essays, I think it's obvious their ability to write at length will increase as time goes on. We've seen this in video outputs, which use to fail at keeping objects consistent for a few seconds and now can do so for longer.
So, I think we still have to wrestle with this basic fact that LLMs allow a full range of human collaboration including 0%.
I don’t think they can write poems either. And they cannot write short stories. AIs will get better I’m sure, I’ve noticed the progress, but as long as it generates rather than imagine, it can’t complete a story.
Well they obvious *can* write poems and stories. Whether they are *good* or not is another question. I agree that LLMs are not likely to create great or innovative literature by themselves any time soon. But most of what people read and buy is formulaic and I could easily see LLMs competing there. Really, they already are. TiKTok and Instagram Reels are filled with AI slop racking up many views and I bet AI generated clickbait websites get decent traffic. A recent study showed AI books getting financial returns in the self-pub Amazon marketplace.
There are already apps with AI stories and millions of downloads. Many (All?) of them are like a series with neverending episodes. And they cost an absolute fortune, you could end up spending 100s to hear the end of a story, a book can be bought for 10-20 or even a lot less. Sad.
We can’t really say that whether they’re good or not is another question. Everyone can write after all and only major writers get that immortal fame.
It doesn’t also matter whether AI generated books bring money. With today’s marketing techniques people are buying virtual things.
We’re talking here about books that matter, with depth, innovative and/or masterful literary techniques. AI is far behind in terms of creating such books, and believe me, I’m a big fan of AI. As I mentioned I do use it and have put together a process that works and helps me advance faster. But left to itself it cannot, as an LLM, create anything that comes close to even an average human writer. Maybe the next General AI would. I’ll try it and will discover. But today, LLMs assemble, they don’t create.
The specific numbers that are worth talking about are the 60/40 mentioned in the article. For people that use AI responsibly, that's not a bad ratio.
The more obvious point: bidding war and many professionals somehow overlooking an AI contribution? If you're so academically inclined, Mr. Intellect, then why do you point the finger at a machine rather than the people disregarding it?
I'll bet 100,000k that all of them knew but the general swing of culture right now is AI = replacement = evil so they pulled their investment proposals to cover their asses.
Let's be real real here. You can romanticize anything you want, I certainly do. But don't sit on your soap box with your romanticized rhetoric. The publishing industry has been shitting on and eating the souls of anyone that's not jk rowling. And now you're protecting them because there's a new "evil" in town?
I'm replying to my own comment because I don't actually want you to go through soap in your mouth. It's not pleasant. But read my comment as something that might be thought provoking. Instead of the human provoking I did to you.
I've tried this as an experiment on a short story. To generate anything without obvious tics and beyond the lowest literary quality is death by 10,000 prompts. It is easier, at least in my case, to in fact write the thing I want generated myself. The only utility I've found in it is in some fact checking and research, in which case it is glorified Google. Caveats apply there, too.
It's way easier to write it yourself, and limit the role of the AI to analysis, critique, proofreading and line editing. With that, I was able to finish a book in 4 months when it used to take me a year.
I'm curious what LLMs are speeding up here. Maybe my process is just very different because I don't think that would speed things up at all for me. Proofreading is a pain but certainly not 8 months time. By "line editing" do you mean you let the LLM rewrite your text beyond mere proofreading?
Sorry I missed this. I'm happy to share with you what I call the ACE method (Analysis, Critique, and Editing). I don't know your personal process; mine is to write, rewrite, and keep rewriting until I think it's done. That's usually around version 7 or 8. Through this AI-assisted technique, I'm down to 4 rewrites.
I write a chapter, run it for analysis, and the AI instantly gives me twelve pages of analysis to ponder. Not to accept, but to think about. Then it critiques my work: what's good to keep and where I'm short on details, where I repeat myself unnecessarily, or where I expand too much on an idea. Then I rewrite the chapter and run it through A and C again. After a few more proposed changes, whether I accept them or not, I ask for editing and move to the next chapter. Once I finish the latest version, I start over.
What you get with AI is immediate feedback; some you disregard, others you use to better your work. At the same time, each version narrows the AI's thinking. Left without the narrowing, it will roam freely in the billions of oceans of information at its disposal, uttering unrelated and sometimes ridiculous notes. Boxing the AI in the book's world led to higher-quality analysis, critique, and editing. It becomes a reader and an editor working beside you at all times, giving you immediate feedback to use or disregard.
It's a fun process, and the fast pace is motivating.
Before, I needed to rely heavily on friends' input or pay a professional - and wait for their input. Today, I'm sending my final version to a professional editor, and it's returning a week or so later with far fewer notes and proposed changes.
I sincerely invite you and other writers to try this method. You write, you decide, you're in control; only there's someone next to you assisting at all times.
Dont worry, youre blocked and I will never read or buy anything you write, editing is one of the most important parts of writing a book, destroying the planet for your convenience is the ultimate in entitlement and indicates that you are not a very good writer.
Considering how many ghostwriters are responsible for NYT Bestsellers and works that make publishing houses $$$$$$$, I'm struggling with the AI debates in this arena. Even high quality AI pieces require extensive skills when it comes to crafting prompts, revising, editing, etc.
*
Edit: Reading through the responses has reinforced that this conversation extends far beyond AI. We've touched on ghostwriters, copyright law, plagiarism, originality, intellectual property, publishing, ownership, and whether there is even such a thing as an original idea. These aren't separate conversations, but expressions of how we've commercialized human creativity.
Human beings are generative by nature. Every company, invention, book, song, painting, technology, and economic system began as someone's capacity to create. Creativity is how we translate our time, attention, experience, relationships, and sweat equity into material value.
Scarcity consciousness (there's not enough for me) shifts our identity away from our capacity to generate value and onto what we've already produced. Once that happens, our worth becomes increasingly tied to external validation of those creations. Publishing deals, bestseller lists, copyright, ownership, attribution, awards, recognition, market value. They stop functioning as ways of governing the commercialization of creative work and become measures of personal worth, turning the artifact into a proxy for identity.
That's why these conversations remain so emotionally charged. A creative work can absolutely be copied, reproduced, commercialized, infringed upon, or stolen. The human capacity that produced it cannot. When we lose sight of that distinction, we begin protecting artifacts as though they contain the source of our value instead of recognizing them as expressions of a capacity that remains inherently human and endlessly generative.
I think in many ways comparisons to ghostwriting is fair, but also... using a ghostwriter is not really artistically respected. It's mainly used by businessmen and celebrities who can't write or commercial fiction packaging companies that often release books under made up names to control the profits.
Yes. Publishing houses had to pivot years ago to start purchasing attention, by way of giving book deals to people based on social media status and reach - many of which require ghost writers.
Then there is gamification of the NYT bestseller list, which is why many titles have a dagger next to them. James Clear dominated with Atomic Habits because allegedly he and his team figured out the precise amount of books that needed to be purchased weekly to maintain his position. Not because it was the best book on the market.
All this to say, crratives need to stop measuring their success/ worth or placing value in these systems and structures because they are inherently untrustworthy.
What authors should aim for: consistent sales over time. That’ll get them much farther than, say, a week or two on a bestseller list & then a drop in sales. It’s also widely known in the industry that the NYT list is curated. That doesn’t necessarily mean the “best” books get on the list. Merit really has less & less to do with being a bestseller. It’s more about what consumers are buying.
A friend of mine was absolutely gutted after their first book went #1 organically and the second got a dagger upon release because their publishing house didn't set up different purchasing portals for bulk pre-sales. I didn't know anything about this system until I started seeing what gets sent to the authors from NYT.
What I’ll say about that is that your friend would still receive the benefits of having a bestseller: store placement, consumer recognition. Most consumers have no idea what that dagger means. They just know the book is on the bestseller list.
and ghostwriters stand in line to be selected..... These writers make oodles of $$ because high-profile books sell and many of them sell quite well....
After stories I've heard first hand about the process of becoming a best seller....never in a million years would I sign those contracts. Literary agents, editors, publishing houses...so predatory. And it works because so many people are seeking validation outside of themselves to feel worthy of their creations.
you make an excellent point….historically, at least since legacy publishers came on the scene at the turn of the 20th century, the ‘gold standard’ was acceptance into this club….we are in a seismic shift, undergoing a radical change that I think those of us in this world lose sight of….I’m interviewing @Kerri Schlottman soon on my podcast…a bestselling author, #literary, who shares her road to publication stories…it’s not easy to figure out and she openingly talked about this at a conference where she keynoted…can’t wait to continue our convo..and will post here…
Your point about ghost writing is on point. What difference does it make how the book was created as long as it's good. The amount of research, input, revising, and editing required to produce something of quality with AI is extensive. Most people who don't use it don't realize that. Also, I put a college paper I wrote 10 years ago into an AI detector, and it said my paper was 78% AI. AI writing had progressed to a point where algorithms think human writing is AI. They are completely unreliable.
Once again, that's the thing. If so much work goes into it regardless, WHY do people who use AI to write their books feel the need to use it? Why outsource their thinking and researching that why? Why not just do all those tasks they're assigning to AI themselves if it's gonna take them sooooo much work and skill input either way?
Agree, the argument is mute. Can one say that cavemen art is the only true art because it was solely done by hand? That a more and more sophisticated brush didn’t make painting more precise, more detailed. I’d like to see anyone here looking at Picasso in the eyes and calling him a fraud, or removing his work from auctions… it’s a shame they pulled a fantastic work because it was made with the help of the newest tool. It’s also a shame that the writer is pulling the race card - but that’s a different subject.
Keep writing and use what makes your writing better without shame or fear. In a few years all this will be in the past and people will be arguing about another silly subject.
The problem with this argument is that it LLMs are a "tool" that can do .0001% of the work or 100% of the work, plus any % in between. This is what makes conversations so tough. Some uses of LLMs are like a paintbrush. Others are like hiring a ghostwriter you collaborate with. And others are basically plagiarism (passing off someone else's work as your own) or fraud.
Picasso used paintbrushes, but without Picasso the brushes and paint would do nothing. OTOH, I can go to ChatGPT right now and say "write a poem" and it will write a poem with zero input, ideas, technique, or anything else coming from me. I just did that now and here is the response
"Where the Light Goes
Morning slips softly through the trees,
painting gold on every leaf.
The wind carries yesterday away
like a letter no one needs to keep.
A river holds the quiet sky,
clouds drifting where they please,
and somewhere in the distance
a bird sings through the breeze.
We spend our lives chasing moments,
afraid that they will disappear—
but some things stay long after,
like sunlight lingering here.
So let the day unfold slowly.
Let your heart be where it goes.
Even when the road is hidden,
there is light beneath your toes."
Would you say this poem is mine? That I should get any credit for it when my entire prompt was "write a poem"? I think 99.999% of people would say no. This is an LLM written poem. The human had no notable involvement. OTOH, if I wrote a poem myself and then asked ChatGPT to offer feedback and it suggested tweaking a line or two, most people would say I wrote that poem.
So the whole issue with LLMs in art comes down to deciding where the line is. When does it go from tool to collaborate to some form of fraud or plagiarism. And yes, it is plagiarism to claim work as yours which you didn't do. And this is not hypothetical. Lots of people are doing more or less what I just did with the above poem and then submitting them to lit mags or uploading books of them to the Kindle Store.
You're all over this comment section claiming that people shouldn't care about AI use in fiction and predicting that, someday soon, nobody will care.
People care. Especially when it comes to fiction. The majority of readers still want to know that the prose they are reading was produced by a human being. These controversies wouldn't occur otherwise.
AI is taking over the creative arts with the same inexorable incrementalism we saw when we gave away our privacy in the name of convenience and safety, but that doesn't mean writers should be uniformly grateful to the tech companies.
We are already seeing excessive reliance on AI in classrooms. Offloading cognitive tasks erodes core mental skills like critical thinking and memory retention. I don't even need to explain why these things are valuable to a writer.
You've claimed here that an LLM can't write a short story. Have you read the short story that won this year's Commonwealth Prize? I'm not talking about copying and pasting it into Pangram. Just read the first few paragraphs. It's incomprehensible. Laughably bad. Yet it hoodwinked the judges.
I don't mean to be so hostile but it's exhausting to see resistance to AI characterized as the ravings of backwards Luddites. There's a bizarre cult-like fervor to some writers' embrace of AI. Any writer who can't see the obvious value of the iterative drafting process that literally IS writing should be demoted to prompter.
Sorry for the typo, I should have ran my reply through ChatGPT 😀
I didn’t say people shouldn’t care, all I’m saying throughout is that LLMs cannot write one chapter of a book and even less an entire literary work. Therefore there’s no need to argue right now. Later, maybe, but right now a writer’s role is very safe.
I can’t speak for judges of literary work who got fooled, their problem, their reputation as judges.
Criticizing the use of AI in education is also moot, the same was said about the internet, remember? How student copy-pasted from Wikipedia or other published works. Still happening among universities’ PhDs btw.
You need mental skills for them to erode in the first place. A degree does not give you that. Today it’s just three words on a resume. Education and Academia, as presented, have failed to produce an awoken class and are now corporative in nature, spreading toxic philosophies and culture wars that challenge critical thinking and are rewriting what memories should retain…
I’m not saying you’re backward, I don’t even think you’re hostile, you’re writing your opinions elegantly and you have all the right to do so.
As I do. I have the right to adopt a new technology that helps me write faster, and better without this “shaming” without thinking. I’m forcing no one to use it, I found a way, using my analytical skills, developing communication methods with a new non-human intelligence to help me write books faster and better.
Many inventions before did the exact same thing (internet, MS Word, a typewriter, the printing press, all the way to the invention of pens and paper) each of these inventions made our work better and faster yet no one argues about them anymore. Technologies have helped humanity, that much is undeniable. And every new technology brought with it a wide range of arguments and opinions, until time proves their worth and their shortcomings. LLMs are the same. In writing, in art in general, they remain tools rather than agents and require the artists’ full involvement, all they do is save time. Kind of like the bean can you simply buy at the supermarket. Does a farmer still do it by hand, caring for every plant, spending days collecting, gathering, preserving, bagging, transporting, selling… should we shame modern beans producers for using all the technologies available? No one is. Why shaming writers who use new tech? Why was a great book canceled knowing too well that the writer did the writers’ fundamental work?
How do you know it's a “great book”? Nobody has read the manuscript. It's not even available online. Are you just assuming it must have been great because of the advance?
The author in question hasn't been able to provide much in the way of working documents, which is odd. Writers love to talk about process. (Probably because nobody ever asks us.) Moreover, no literary agent would throw away an advance that big without good reason.
This “it's a tool, just like any other” refrain is either a canned speech or willful blindness. For all of human history, tools have been passive extensions of the human body, like a hammer (hand/arm) or a camera (eye). AI actively generates content, makes decisions, predicts outcomes and allows humans to offload cognitive tasks. It's much closer to an autonomous agent or digital worker. It can used as a tool but it's not just a tool and repeating that same line over and over won't make it so.
Writing is more than merely generating prose. It's a process. A valuable one. Outsourcing that process because you're busy and it's faster misses the point, IMO. Writing is about quality, not quantity. Churning out books faster does not mean they're better.
Yep, you're free to use AI. Just like I'm free to think it sucks and writers who adopt it wholesale are deliberately selling themselves and their writing short. But demanding that everyone else get on board while making the same tired points about it's “inevitability” is like throwing your lot in with Goliath and wondering why David doesn't like you anymore.
You don't get to use a technology that does research, edits, and writes for you AND demand respect for that choice.
A book that gets a $2m contract from publishers is usually a very good book. And we've heard your refrain over and over whenever a new tech takes its first step. All your purity arguments fall apart each time, and the technology wins the day. Then you complain about being left behind.
You're putting words in my mouth that I never used, exactly like someone who can't hold a serious argument. I don't care about your respect, and even less your opinion once you start playing these losers' tricks. Go bore someone else please.
From what I understand, via research, is that you cannot copyright AI content. All AI content falls in the public domain. This means if a publisher invests a lot of money into a book, once it is published anyone can copy it and sell it. Without a copyright, the publisher can’t do anything about it. Who is going to risk two million dollars or whatever when this is the case? If you study copyright laws as of today, you can copyright the human parts but not the AI generated parts. So, even if two percent of the book is AI, you can’t copyright it as a whole.
How in the world can anyone ever know whether it was AI or not? Do you know that a very well written piece using proper grammar and good sentence structure often comes up as looking like it was written by AI. It's time to get over it and on with it because it's here to stay.
According to the copyright office, they can. I only looked into this as I use AI to assist me and wanted to be clear as to what I can and cannot do should I ever want to publish anything. Personally, I could care less if people use it or not. Actually, I am glad some people do as there are people with brilliant ideas who cannot express them properly in writing.
You should look at the actual law. Material created with AI support is copywritable unless something has changed in the past few weeks. AI is presently seen as hired help. Many authors employ teams of writers already and have been for decades.
Thanks. I have dug a little deeper to understand what constitutes material contribution. Plot development, character definition, world-building, all work. And in fact AI support for world-building is very helpful.
No, hardly. The value of AI in world-building lies in consistency checking, inconsistency discovery, alternatives evaluated against proposed new mechanics.
I suspect you think AI writes novels, I disagree. AI, by itself, does absolutely nothing. Someday, we may be able to task AI with creating a book with the prompt “write a good book”. That would not work today. But if it does materialize, when it does, resistance will be futile - but there are other responses.
I think you're assuming AI is only functioning like knock-off factories. It's not. Human also absorb ideas and concepts their entire life and create from experience. This is why I'm fascinated by the human obsession with ownership and control - especially in creative fields. We never truly own anything. We never truly create anything from scratch, but our egos are developed through serious lack and scarcity consciousness, which perpetuates a deep desire to death grip what we 'own', instead of embody the essence of how we created in the first place.
What makes AI dangerous is it is entirely a product of and servant of corporate capitalism designed for generating corporate profits. Its incentive structures and whatever legal and political structures develop around it will be designed to protect its value to privately owned corporations at whatever cost and future blight it has for the great majority of human beings and for human culture into the future. Whatever tiny shoots of human artistry, community or creativity may appear will be crushed or turned against humanity by the overwhelming and deadening force of corporate capitalism. Look around if you think that's not true.
That and the fact that it appears to reinforce a false view of the self as a sovereign entity. It is a form of power. As such it will be unevenly distributed. That uneven distribution will further the divide between the ultra-rich and every one else. It is the ultimate extraction technology.
What should be the proudest moment in human history, will actually be the worshipping of an AI idol, and some severe corruption due to the further increase in power inequality.
That is the real problem, but there is hope. Visit my page and read “Is the Internet Eating Itself.” Warning, you have to have an open mind in relation to astrology
There are already several copyright infringement lawsuits in progress from several content providers who are suing AI companies for stealing their content.
A LOT of AI content was already stolen and violated copyright of the source providers. It is well known and documented that AI companies scrape content without regard for copyrighting.
Not metaphorically. Not poetically. Not in some abstract, academic sense that can be debated in conferences and footnoted into irrelevance.
You have robbed us—clean, deliberate, and systemic.
Our minerals.
Our natural resources.
Our history.
Our heritage.
Our arts.
Our craft.
Our stories.
Our religions.
Our legends.
Our myths.
Our tales.
Our words.
You took what could be carried. You extracted what could be sold. You catalogued what could be studied. You renamed what could be owned.
And what you could not absorb—you vilified.
You called it primitive. Pagan. Backward. Dangerous.
You erased it from classrooms, mocked it in pulpits, outlawed it in policy.
Then you wrote books about it.
Then you copyrighted those books.
Now, after the heist, after the archive, after the rebranding—you turn around and lecture me about ownership.
Copyright.
You say it like it is sacred. Like it emerged from pure intention. Like it was not built on top of everything you already took.
And now, we arrive at your newest anxiety: artificial intelligence.
You are outraged that it has been trained on your data. Your books. Your images. Your systems.
But let’s be honest about the lineage of that data.
What exactly is “yours”?
The stories shaped by structures you inherited from us?
The aesthetics influenced by forms you extracted from us?
The knowledge systems you once dismissed, now repackaged and monetized?
You are worried about theft now?
Interesting timing.
Because from where I stand, AI is not the beginning of the problem. It is the exposure of it.
It is a mirror held up to centuries of accumulation without accountability.
So no—I will not perform outrage on your behalf.
I will use it.
Unashamedly.
I will use it because my fingerprints are already in the foundation of what you call “your” data. Because the sweat of my ancestors—and my own—has already been absorbed into the systems you defend.
And if I cannot use these tools to build something entirely new, then I will use them to dilute what you stole.
Authors have already been doing that for a very long time. Take John Carter to Flash Gordon, to Buck Rogers, to Adam Strange, are they not all basically the same (different names, different environments - if there are only 7 to maybe 36 fundamental main story themes that can exist, why question originality now)? There is a confusion in comparing one timeline of literature to the current timeline of literature. If, say, Star Wars and or Harry Potter were released now, instead of then, would they be in dispute (potentially even algorithmic flags for plagiarism)? Is not every story built on top of every other story? What is then being potentially judged is an individual's knowledge about history and or a variety of disciplines, and can then connect all those dots into a pattern that works (intuitive thinking and cultural literacy). Or how one writer might seem cold or clinical to another who expresses a more emotional lived experience. How do you judge that, who gets to judge that? How do you judge where one person gets inspiration and turns that into a creative thing, over where someone else gets theirs? We are allowing the existence of “AI” to poison the well of perceived human effort. Not to mention every program or platform has some form of “AI” (which it is not artificial intelligence; that does not exist): the entire Adobe Suite, all translating programs, and most, if not all, editing programs use it. So then what are we talking about: some form of writing orthodoxy (only handwritten manuscripts, and a lawyer, or someone from of a legal witness, monitor, from when you have that first spark of an idea to the finished text?). Side note: you cannot copyright the entirety of any given text; language in and of itself is not owned, so you could only have some form of ownership of specific sentences that are inherent to the context of that text (a sentence with a specific character’s name or the name of a place specific to that storyline), but you can not copyright the generic sentence: do you want to go grab a cup of coffee.
That is bs on it's own, when Mfg can manipulate a finished product by more than 30% I believe and call it an original. Same with imported products into a FTZ. To have such a stringent REDICULOUS rule (...even 2% is too much) when AI assist has been shoved down everyone's throats as the 'go to' then turn around and severely penalize usage? Yeah a monkey came up with that logic. Suggest to push back the bi-polar ruling.
On the other hand, I like the idea that AI assisted works, are not original. It seperates the posers from actual disciplined daily/regular writing. Edith Wharton, guilded age author.
AI assistant? Means Any writing or design of original work; enhanced, edited, co-author, relied upon for content = eventually Not an Original work. My strong but humble suggestion is don't give anyone a reason to claim your work. Refrain from using. Fk them.
How is using AI to edit work, any different to using a human editor or a thesaurus or spell-checker? As long as you use AI to mark the suggested edits, and you accept, reject or modify them, then I don't see that as an issue.
Excactly! That’s actually the loop hole. As mentioned if AI is used to ‘create’ the text, subject, storyline? That’s when this $2M deal gone wrong applies: Actual copyright - however, the laws have NOT been updated to show ‘who’ would be the copyright owner of AI generated text? That’s the dark hole real question of who owns the work? And who get’s for it.
Probably fake academic work takes a little effort, too. Or fake Rembrandts. Or laboring with love over a fake wine label to create the aura of age. It’s about integrity.
He didn't copy anything. It was not a fake Norman Mailer or fake any other name writer. He used current tools, like the people before him used word processor and before that type writers.
But it is. The entirety of AI's knowledge is made up of stolen art. It's the work of all the Mailers and all the Rembrandts combined. Same with AI music. It's stolen from everywhere and everyone.
According to your interpretation almost all PhD dissertations are stolen from everywhere and everyone because they are based on preceding knowledge and studies of books in the libraries by other authors. If ingesting existing work to learn structural patterns, stylistic techniques, and foundational principles constitutes "theft," then every scholar, painter, architect, and writer in human history is a thief.
They are inspired by everything they come in contact with. Much of the tim we are impressed upon subconsciously and what we believe to be original isn't completely original. How do you xplain multiple people having the same independent idea? Liz Gilbert talks about her own theory on how she believes ideas are formed and what happens when we don't act on them.
Just like graphic design and digital music, AI is absolutely pressing on the starving artist complex and exposing the lack and scarcity wounds in creatives. Creativity is infinite.
AI is a tool that still have many misconceptions because more people have opinions about it, than experience using it.
Ok, let’s take music as an example; in the last 126 years, as a very conservative estimate, about 100 million songs have been recorded, with the same variation of 4 chords. Yet, by any measure of legal, artistic, and or cultural standards, almost none (only an incredibly small fraction, a few hundred maybe, out of millions) of them are direct copies. So then what is, or should be, the boundary between a shared cultural structure and stealing a work? And who judges that?
That's not at all the same though, is it? All those people who used word processors and typewriters still chose every word themselves, agonised over structure and meaning
But, did they? Did they run their ideas by a spouse or a friend who may have made suggestions or recommendations or came up with plot ideas that they ran with? My dad constitute an exception to having their own unique ideas? Would that disqualify someone from publishing a book? I think it's just all been blown out of proportion.
People all work differently. But most fiction writers or poets--non-fiction might be different with research assistants--do write almost everything themselves even if we take in feedback from friends or editors/agents. Those people often provide open ended feedback, not rewriting the sentences, and the final decisions are the authors. If you looked at the author's final draft before editor and agent feedback and the final book they would typically look very similar. Many of my short stories have been published with no one else looking at them except the magazine who might offer nothing more than a few typo corrections for edits.
That's quite different from having an LLM do major rewriting. That is far closer to having a ghostwriter or writing in collaboration with another author.
I don't disagree with this at all. But it is interesting that we sometimes get significant input from other people, with whom we may discuss our ideas or show them a draft of the manuscript and they may come up with great suggestions, and I don't think we always credit them as contributing authors or anything like that.
Seems extraordinary to me, the lengths to which you and others, like “Gene”, below, are going to legitimize this scheme by which a person claims to have written something that he did not write, to have created something he did not create.
But everyone wished to have bought a work of Andy Warhol.... Warhol literally declared "I want to be a machine" and built The Factory around silkscreening, mechanical repetition, and assistants executing his concepts. His entire career was a deliberate challenge to the romantic obsession with manual, unassisted "authenticity." The Cretaceous period is long gone and we have been privileged that the tedious work is getting less and less. Assuming that somebody using AI as a tool did not write a book,and did not create what he wrote shows a fundamental misconception how LLM actually work.
But this is a kind of myth, no? The writer, the artist, in the studio, isolated from the world, an island unto themselves, the lone creative, and all credit is somehow theirs. Let’s take an architect, for example; they do all the initial research (where does all that research come from?) and drawings (is that style and understanding of drawing solely unique to them and only them), then there is the construction of whatever they have designed: do they do any of the labour - the carpentry, masonry, plumbing, electrical, roofing - do they do any of the manufactioning and or the production of any of those materials - did they invent those concepts and or products: plastic, ruber, the boiler, joinery systems, how and where things are mined, but in the end their’s is the name on the building. There is always a mass of invisible labour, and a supply chain behind any idea that gets produced. Or, Andy George’s six month experiment to produce a chicken sandwich from scratch.
No. Try to throw a chapter of say, Camille Paglia's Sexual Persona into chat GTP and it will come out as non offensive, nobody-get's-hurt kind of compliant drivel without bite and the polemic removed. Most probably chatGTP will refuse to work on it because the content "may insult" or, oh my God, "hurt other users". For example, I threw Paglia's infamous NYT op-ed of 1991 "Ninnies, Pedants, Tyrants and Other Academics" into chatGTP (without the authors name) and prompt it to "correct" it. It refused: "This content can’t be shown because it may conflict with our guidelines for creative content."
Unless you write a book which conforms to corporate compliant language rules and guidelines (like the drivel on Linkedin) you won't be able to have AI write anything with an authentic voice. It will smoothen the edges, hedge any plain statement (one view to many views) and throw in all the ideological veneer of the last 15 years. Values which America assumes as usual are universal but which come across as colonial if you live in a civilization with different value systems. A kind of white man's (and ironically now black man's) burden in a new modern coat but with the same colonial whiff and a version of Newspeak of Orwell's 1984.
American identity politics, modern diversity rhetoric, and corporate "alignment" rules have become the new moralizing export in all US produced AI platforms. When corporate institutions enforce these sanitized standards globally, they present themselves as progressive, yet they execute the exact same patronizing policing—deciding what language is "safe" or "acceptable" for everyone else.
If any random person uses an LLM, they get generic, standard output, sanitized, risk-averse corporate noise, because they lack vision, technique, and editing discernment. The quality of what comes out depends entirely on the sophistication, direction, and critical eye of the person using it.
I doubt fake works will slip by editorial teams. Any serious academic is known by colleagues in the discipline, and academics usually works with multiple authors.
No more fake than artists tracing or projecting photos to draw. It used to be a big no-no. Now, no one cares except the most obsessive old-school art teachers. Renaissance and later painters supposedly used optical devices to help with perspective; Vermeer was frequently accused of doing it. You can argue it was the technological equivalent of using LLMs for writing today. Do we care now?
This is exactly where I struggle with the AI debate.
I recently ran one of my own articles through an AI detector out of curiosity. The experience was mine. The thoughts were mine. The photograph was mine. I wrote the original piece and used AI to help edit it.
The result? **0% human.**
Apparently, my thoughts + my experiences + my writing + AI editing = no longer human.
Meanwhile, publishing has used editors, ghostwriters, researchers, and entire teams behind authors for decades.
I understand the concern about someone having AI generate a book and passing it off as entirely their own. But if we're going to start judging authors by unreliable detection software instead of the actual creative process, we're heading into some very muddy territory.
It’s both comical and ridiculous: AI is now supposed to decide what is AI - a snake eating its own tail. And if AI can supposedly detect AI-generated text, it’s not much of a leap to have AI generate text that the same detectors won’t recognize as AI-generated. This makes the AI detectors a naive and temporary idea.
BTW, these are LLMs. There’s a perfectly respectable view that they aren’t really “AI” yet, but extremely sophisticated statistical language-generating machines
- Security cameras tracking your every move (AI recognition, behavioral analysis)
- Your loyalty card tracking exactly what you buy, when, how often (AI profiling)
- Store owner's AI predictive analytics about your shopping patterns
- Facial recognition at entry (some retailers now)
- Behavioral tracking (how long you stop at each shelf, what you pick up and put back)
On the way back:
- All the same stuff again
- Music streaming algorithms (AI)
- Targeted ads following you around (AI)
You probably touched AI 50+ times. Conservatively. In one errand.
Half of it is *surveillance*. Your bank analyzing your behavior. Street cameras tracking your face. Retail stores mapping your movement patterns. Your loyalty card building a profile of your entire life. AI systems making decisions about you based on all that data.
Nobody's panicking. Nobody's calling for authentication. Nobody's saying "prove you actually wanted milk and not AI milk." The grocery store isn't holding an emergency board meeting about whether your purchase was authentic.
The bank isn't issuing statements about the "integrity" of your transaction. The surveillance system isn't requiring you to prove you consented to being tracked. Street cameras aren't requiring verification that you're "really" walking to the store and not some AI simulation of walking.
The Actual Situation
You live in an AI-saturated world. Have for years. It's so embedded you don't notice it.
But the moment you use Claude to tighten the prose in your cookbook, suddenly it's a moral crisis. Suddenly we need industry standards and authentication and Pangram scans and public accusations.
The publishing industry didn't get angry when their distribution networks went AI. Didn't protest when Amazon's recommendations went AI. Didn't lose sleep when their own PR platforms switched to AI. Didn't demand authentication when grocery stores started using AI pricing algorithms (which, by the way, are arguably more ethically fraught than you editing your manuscript).
They only get angry when *authors* get access to tools that used to cost $15,000.
The Actual Moral Hierarchy
Here's what's real:
Surveillance AI (bank analyzing your behavior, street cameras tracking you, retail mapping your movements): Invasive. Ethically questionable. You're a subject being profiled. Your data is being sold. Your patterns are being predicted and monetized. This is actually worth panicking about.
Your cookbook being edited by Claude: Harmless. You're using a tool. Your work is still yours. Nobody's profiling you. Nobody's monetizing your surveillance data. This affects nothing except your editing costs.
The publishing industry is fine with the first (doesn't even mention it). Panicking about the second (calls emergency meetings, pulls book deals).
That tells you everything about what they actually value.
What This Really Means
The panic isn't about AI existing. AI has been doing the heavy lifting in publishing infrastructure for years. They've been fine with it. Profitable with it. Meanwhile, they're completely comfortable with surveillance AI that's infinitely more invasive because it doesn't threaten their control.
The panic is about *visibility*. About the fact that you can now see the tool being used because you're the one using it. About control fragmenting. About economics shifting.
So they manufacture crisis language:
- "Authenticity
- "Integrity"
- "Readers deserve to know
- "Industry standards"
- "We need to protect literature"
Translation: "We need to stay in control of what counts as legitimate."
The Test
Ask them this: Remove all AI from your publishing operations. All of it. Your PR platforms, your contact systems, your recommendation engines, your distribution networks. Shut it down. Do it today.
They won't.
Because they don't actually believe AI invalidates work. They believe it invalidates their gatekeeping power.
The Final Hypocrisy
They're using Pangram to accuse you of using AI.
Pangram is AI.
They're using an artificial intelligence system to detect artificial intelligence use, calling it "proof" that your work isn't authentic, while they themselves remain completely comfortable using AI at every level of their own operation.
They won't give up the surveillance AI. Won't give up the profiling AI. Won't give up the infrastructure AI. Won't give up Pangram, the AI detection tool itself.
But you should give up Claude. For integrity. For authenticity. For literature.
Drive to the grocery store. Notice everything running on AI. Notice nobody panicking. Notice it's working fine. Notice the only people asking you to sacrifice anything are the ones using more AI than you've ever touched.
That's not a crisis about literature.
That's a tantrum about losing gatekeeping leverage, dressed up in moral language and powered by the very tools they're accusing you of being corrupt for using.
Agreed! It's the death rattle from an industry that can no longer buy the attention individuals can generate on their own. They started offering book deals to anyone with an audience because they cannot compete in the attention economy .They literally have to purchase their audience, NDA the hell out of the people who bring the audience, and buy their IP/Name/likeness etc to profit in perpetuity. It's gross and the individuals inside that industry need to re-purpose their skill sets into something generative instead of extractive.
I think you're correct that LLMs let people generate their own content and self-generated content dominates the current "attention economy." This is a big threat to AI-assisted writing too. Why pay for content someone else didn't write when you can generate your own tailored to your specifications?
Every social media influencer above a certain amount of followers is being offered a book deal. Steven Bartlett has pivoted his media company to signing deals with Podcasters that already have established audiences and streams. Companies are not buying content, they're buying eyes in they same way they're buying data.
Anyone seeking external validation and who believes they need someone else to prop them up to be 'successful will sign these inherently bad contracts.
Everyone thinks the AI conversation is about creative agency when it's really about self worth. Creatives have notorious chips on their shoulders about their creative worth, which is why the 'starving artist' mentality persists and why many will continue to suffer for their art and feel victimized by this technology.
The question here that sorts all of this is: do you have a non-ai, human centred process you follow to produce your work? If you do, then in the production of your work, as long as you only use AI to augment your process, not replace it, then that means the work is still yours.
I so agree with this and for people who are really doing their own work and then using it for editing. The programs that read writings for AI content have some issues as well.
Hi Nicole... This conversation definitely extends beyond AI. The Tech position/push is for ...fair use... as purportedly, despite the strong use of AI to create a piece of art, AI produced work is being defined as belonging to the public domain. I am not certain if I agree with this legally, as the very definition of copyright applies not to ideas, but to the expression of those ideas. Trillions of bits of data that belongs to us all is the fuel that AI programs use to produce creative works.. The content belongs to everyone, and this is an issue that demands our attention. Without our input, there is nothing for AI programs and systems to use. This raises several issues, and we should explore how contributors to AI can be enriched by making possible the creation of these works developed/interpreted by AI programs and systems.... Who knows... Maybe we can create an economically just society.
Falade, lost me at “Three Black authors received major publishing deals this year, and all three later had deals cancelled or disrupted following suspicions about AI.”
If there ever was a time not to be played the race card this is it.
You’re telling me that the alleged-ai-assisted authors “used the spectre of systemic racism”???
My point was that there’s a very good chance that non-poc authors have been using ai and getting away with it, and that they weren’t caught because of their privilege and the default level of trust granted to them and not granted to people of color.
Yes, Falade did exactly that. Did you read his response?
Listen, in many aspects of American Society POC certainly face discrimination. I don't deny that, but in both publishing and academia, they are often given much greater deference because everyone wants a diverse roster and nobody wants to be accused of racism. You need to realize that some environments/disciplines become ideologically captured and the larger power dynamics get interrupted and sometimes reversed.
I think both black and white authors are probably using AI unethically and using it at the same rate, but there is not anywhere near enough evidence to prove that racism is the cause of this publisher pulling the book.
According to your flawless logic, catching someone red-handed isn't actually about what they did, it’s just a tragic case of unequal surveillance.
Apparently, when non-white authors get caught using AI, it’s not an ethics violation; it’s a systemic injustice because we haven't caught enough white people yet! (Am I understanding your logic"?)
Your defense: "Sure, my client cheated, but what about the hypothetical white people who might be getting away with it?"
You're lauighable logic is a masterclass in shifting goalposts! As for your tearful claim that white authors are floating through the publishing world getting "showered with praise" for dropping AI prose into their manuscripts, here are 3 reality checks for you:
Mia Ballard: A white horror author who's novel "Shy Girl" was canceled by her publisher Hachette following an investigation into her use of generative AI. Her publishing career was essentially nuked overnight. Where was her magical "privilege" shield?
Stephanie Johnson: Disqualified from the $65,000 Ockham Book Awards because AI was flagged on her book's cover, a cover designed by her publisher's design studio that she had no part in creating. Total zero-tolerance, immediate disqualification.
Elizabeth Smither: Was stripped of her eligibility and booted from the exact same $65,000 Ockham Book Awards for the exact same reason, an AI-generated cover image she didn't even design.
Turns out the publishing industry is nuking anyone caught using AI, regardless of what skin colour is typing. But please, keep weaponizing "systemic racism" to defend literary fraud, it’s much easier than admitting people should just write their own books.
Universities have actually stopped having students run papers thru AI checkers because they flagged papers that students could prove were not AI generated. If I’m writing now I’m saving every 5 minutes and emailing copies to myself so I have receipts
Unfortunately not enough. Saw on reddit how many lost their scholarship or was downright kicked out or nearly kicked out because of that dumb shit. How are you gonna up and accuse someone of using Ai while USING AI TO SEE IF AI WAS USED OR NOT?! They even said at this point you have to literally freaking record yourself writing and typing out everything and still probably won't be enough.
My god please dont be someone who has a strongly preferred way of typing or have aspies, otherwise, your done.
Even with the written manuscript, notes section with your jumbled up notes and a new note with the story chronological in order and a final typed out version that's done after your done writing the manuscript IT WON'T BE ENOUGH. All because of some bitter lazy cunt.
I graduated from grad school in 2017. I used Grammerly as a thesaurus but it would pop up and ask if I needed help. I never took it as I’m a strong writer but could someone be expelled for saying yes?
We were supposed to submit papers to be checked for plagiarism but most professors refused. The experienced ones always said they learned students authorial voices very quickly, I remember one prof saying within the first week that because of my large vocabulary (I’m still not sure how she knew that based on a few classes) I was most likely a good writer. AI is making us doubt our own brains
On my research I saw a number of them got in trouble due to using it or others similar to it – But it was when they plugged that Ai bs into it or was waaaaaay too formal and proper. I wouldn't use any of those assistance texting because it just adds to them claiming you used it. Type it out leave and have a billion errors and go back and then clean it up manually – Yes, it'll take a billion years, YES it will drain you out as i do when I go and have to redo my short stories and novels and screenplays, but its so worth it in the end and helps the flow in your brain btw.
Plus also noticed, the more I use spellcheck the more I forget how to spell and use words and punctuation which weirded me out BIG TIME.
And I saw a video on yt this girl called readers out on throwing tantrums when writers use big or unfamiliar words; They wouldn't last a few minutes with a Cormac novel or reading PUSH. You can't use big words, can use these ( – or – or —) without some idiot screaming AI! AI! AIIII!!! Cant be expressive and highly descriptive then getting right to the point like NCFOM – Like we can't do anything without some idiot saying its Ai. Remember the good old days when your sole worry was someone saying you stole your idea or work from someone else or ripped off so and so.
And sounds like she was either sizing you up or maybe you were one of a very few to not only do the work but love it and not complain or half ass it? Its so stressful it really is. Gotta basically censor ourselves out of fear of one idiot accusation without proof. They already don't read our manuscripts or assignments or anything as is; Be lucky is they skim through it. Don't know how they can wake up and call themselves teachers or publishers or editors without even looking at the damn work we produce but come to a quick; And usually incorrect. Conclusion.
Its either dumb yourself down and don't stand out or... be accused and ruined for life... Sorry for the rant btw broskie.
I hear this argument a lot: "How are you gonna up and accuse someone of using Ai while USING AI TO SEE IF AI WAS USED OR NOT?! " and it makes zero sense to me. Students are demonstrating competency when they write a paper. Professors are not demonstrating competency when they evaluate the papers! They are trying to determine if the student has met the objectives of the assignment and part of that is making sure they don't cheat.
You missed the entire premise: when a student uses AI to write their paper, they aren't demonstrating competency at all.
An assignment is supposed to be evidence of the student's internal understanding and critical thinking. When an LLM generates the arguments, the paper proves nothing about what the student actually knows.
Claiming a student who used AI "demonstrated competency" is like claiming someone ran a marathon because they took an Uber to the finish line.
Eh, it fits the same mold that racial prejudice can be sneaky. You don’t have to be overtly racist to scrutinize people differently without realizing it. When you see a pattern of black authors getting x 80% of time and others getting only getting y, that could be why. However I agree three is simply not enough datapoints to jump to that conclusion and that makes it feel like card being played to rally people to their defense, not a real accusation with evidence and merit.
(a) Falade doesn't know how many Caucasian, Asian, Indigenous, green, and purple people have been rejected or had their advances rescinded because of AI use; hence, to your point...
(b) Without a full picture of the publishing industry's purity testing of manuscripts, Falade jumped to racism, likely in an attempt to shame his publisher and garner public sympathy.
Falade's accusation of racism is a blatant deflection. By substituting outrage for evidence, he undermines his own credibility. He needs to do the hard work of proving he wrote his novel without AI, rather than hide behind unfounded claims. Crying "wolf" only trivializes real discrimination.
Good thing you only speak for yourself, but I gotta hear this… please explain why you believe “race card” is used by racists and provide the data you used to come to your belief.
If you spent any time on the cesspool X or see any racist Republican talking points, you'll hear it at some point. I think you know this, and you're just trying to be coy. If you don't know any of this... where the hell have you been? I suggest you do your own homework for the data.
Ah, the classic battle cry of someone whose entire argument collapses the moment you ask for a single, solitary source.
Congratulations, by the way; you’re officially the first person I’ve ever seen attempt the intellectual gymnastics of claiming the "race card" is only played by racists. Masterful projection. Truly, 10/10!
*I commented this further down on a similar reaction but felt it was a valid take to repeat once more if nothing but for a different perspective*.. You don’t have to “race-bait” to question any repetitive accounts of a similar nature. And news flash,people, like yourself are not easily lead to believe that someone is being discriminated against,especially for being black in anything! This current climate of perception vs reality in regard to racial biases provides not a single black artist, author etc any favors. In fact,as evidenced even within your own response, it elicits more scrutiny and negative reactions than anything with empathetic understanding. I do not believe this author was trying to use an imaginary “race card” here, but was concerned with his own findings and like most black people in general, truly hope and even fantastically wish that to not ever be the truth. Remember, just as this author openly questioned the process based upon his own meetings with everyone involved,you chose to question the author’s validity based on your own limited understanding of what he was told behind closed door meetings where transparency rarely exists beyond what is recorded.
You conveniently ignored “Falade doesn't know how many Caucasian, Asian, Indigenous, green, and purple people have been rejected or had their advances rescinded because of AI use.”
The publisher was willing to pay $2 million to a black author, but now that Falade’s use of AI has been questioned BY HIS REPRESENTATIVES, NOT THE PUBLISHER (Funny how everyone missed this fact - Read the article; it’s clearly stated in the 2nd paragraph.), it’s a “race issue” rather than an issue of “AI usage”?
Explain what a person’s skin pigmentation has to do with their being caught doing something they shouldn’t have done.
Is the issue at hand a race issue or an AI usage issue?
I find it most telling when someone defaults to the “ I don’t see color” or “green or purple people” explanation.Let’s be clear here, there are no imaginary green or purple people. We have what we have and that’s diverse enough without making things up to express a lack of intolerance towards a specific race/culture. I have no idea what the author knows in regard to how many other authors, and their ethnicity have been issued the same treatment. And I’d gather that neither do you.You are simply adding your opinion based on your experience(s).And I shared my perspective as well. But let’s wait to see if all the details come out first before we write someone off or show callousness towards how they present their story. The same goes for his reps.I can’t say they were right or wrong in the situation yet,so I won’t judge either way!
Appreciate the zoological update on imaginary colours, followed by a full essay dedicated entirely to saying "I don't know."
Here's what you do know: Falade's agents walked away from a guaranteed $300,000+ payday.
Based on my experience, people don't walk away from $300K without a very good reason; hence, Falade's agents walked away knowing, with absolute proof, that their client had used AI to write his novel.
Nobody cares what colour Jerry Falade is; it's the character underneath that makes him so disappointing, as you are also trying to excuse someone's lack of ethics because of their colour.
One would reckon that if, as the trifecta of Falade, HM Wolfe, and Mia Ballard have proven, if it were not at least partially racially motivated, there would be more public news stories, if not more private industry chatter that was worthy of industry-press reportage, about white writers losing opportunities or having already published work cut for the same, especially white authors who have admitted to its use several times. To wit: if it were in fact at least aiming for parity, James Frey would have lost out on the opportunity to publish *Next To Heaven* at least in 2025, if not in 2023 when he said verbatim in an interview that he doesn't care how readers view AI use in his fiction.
The public narrative isn't one Falade constructed. The mere fact that publicly only Black writers have not only been punished but had their punishment publicised would still be an issue even if one can prove parity, because then the question would be why the industry, if not the press, would not be concerned with portraying the issue equally.
Brandon, conveniently glossed over a massive detail: the publisher didn't cancel the deal, Jerry Falade's own agents voluntarily pulled the manuscript and fired him as a client.
Marc Gerald and Sandy Hodgman willingly walked away from a guaranteed $300,000+ commission on a $2M+ multi-publisher bidding war because, in their own words, they could no longer "authenticate how the manuscript fully evolved" after Falade’s story about his process kept changing. Agents don't burn a seven-figure payday on a whim unless there are red flags are glowing in the dark.
This isn’t a unilateral industry ban; it is fundamentally an AI usage and disclosure issue. Every publishing house, and every literary agency, has a completely different threshold for what level of LLM involvement they tolerate, ranging from "zero-tolerance" to "fine for research, off-limits for drafting."
When your own literary representatives pull your book from an active $2 million auction because they can no longer vouch for its authenticity, the problem isn't the publisher's policy, or some "ism," it's the origin of the manuscript.
That detail is not relevant to the argument I am making. It's kind of rich to say I am conveniently ignoring something when you didn't respond to the core of my message.
To put more bluntly: if Falade *did* in fact fake the text by AI or other means, he does deserve to be punished; and if his agents were gun-shy for any other reason, then the specific redress issue Falade has is specifically with them. The individual agents representing anyone are allowed to say no for any reason, that's how the industry works; and Falade is welcome to defend himself as anyone else would be in this instance, so the only part that matters is the evidence himself — evidence Falade insists he has and they never asked for, which is why I imagine he's planned to take it to court.
But it is undeniably still noticeable that consequences like Falade's, regardless of the arbiter, are based on assumptions of dishonesty that have only been publicly documented and discussed when it affects writers of colour under suspicion, when white writers who confess to the same deed we agree deserves consequences face none.
What part of “his own agents pulled the book from a $2M auction” isn't sinking in?
You really think veteran agents walked away from a guaranteed $300,000+ payday without absolute proof Falade used AI? That's not how commission-based businesses work.
And spare me the whataboutism. Playing "Well, what about other writers?", which is childish, completely ignores the fact that every agency and publisher has its own distinct AI policy, from strict zero-tolerance to research-only. Falade was held to the rules of his agency, not a fictional industry-wide standard.
Bingo. Today, what professional in publishing or anything else wants to be accused of racism? Are you kidding? Those folk did backflips and other contortions before they brought themselves to withdraw a black author’s submission AND forego their fee.
Since you don’t speak for everyone, just yourself, using the word “we” is a lie.
If you’re going to be easily offended, bitter and angry then at least don’t lie and understand you only speak for yourself.
P.S.
The topic at hand is about AI usage. Trying to make this about race is nothing more than gaslighting to deflect from the Falade having used AI to write his novel.
No, I said what I said, we don’t care. And no amount of name calling is going to change that. Without knowing any facts of the wrongful accusation against this author, you’ve already written it as a fact that he did use AI. As I said we don’t care.
You can stomping your feet in anger all you want, you won’t change the fact you only speak for yourself.
Can you read?
THE AUTHOR’s REPRESENTATIVES, NOT THE PUBLISHER, pulled the manuscript, therefore in all likelihood Falade used AI. (This is clearly stated in the 2nd paragraph.)
In any case you are obviously bitter and angry and rather stay that way by looking for unsubstantiated reasons to stay that way as opposed to staying on the topic of AI usage, which has nothing to do with skin pigmentation.
I agree. I felt the same. This is about AI use in claiming ownership of one's writing. Why did the race related assessment get in here, and if that's a valid observation, then that's just very sad state.
Some research does show that AI detection software is pinging non-native English speakers, writers of color, and neurodivergent writers at higher rates than native neurotypical white Americans.
It's not just a black issue - it's a problem for many minorities.
I guess you can find excuses using “some reserach” to try and justify unethical behaviour.
If you read the article you then know that it was his agent who withdrew Falade’s manuscript, not the puiblisher. Therefore, if his agent didn't have proof that AI wrote part of Falade’s the book, he'd never have walked on his 15 - 30% or whatever a literary agent receives.
This isn’t a race issue, this is an intergity issue.
Regardless of who pulled it or why, (and yes, I read the whole article?), if it was because an AI detection software pinged some of it as AI, there is plenty of research to support the prejudice of AI detecting software. Of course "some research" is more correct than "all research". Very few things are absolutely conclusive. An agent calling a false truth does not make the truth any less false because it was not done by a publisher. They both have a lot of money to gain and reputation to lose. The unknown: is the agent inadvertently lying in an attempt to avoid controversy that may lead to even larger losses, or is the author lying? We would need to know what the evidence was, how it was gathered, and determine how often that method fails in independent testing surrounding prejudice. I tend to trust a human over AI, solely based on the vast failings of AI. AI is great at managing data, but not so good when it comes to "forming conclusions" or "creating". Using AI to detect AI seems incredibly fallible on its own, due to a lack of objectivity, independence, and neutrality. Have you been following the research and testing of some of these AI programs? The data says it all - they are incredibly fallible, and on some cases outright dangerous.
Groundbreaking analysis, Tina. So because AI detection software isn't 100% flawless, an agency isn't allowed to protect its own reputation?
I guess that if some research says AI detectors are bad, then by all means, let’s completely disregard Falade’s now-former agency’s right to make decisions for their own business. Because clearly, an agent making a business call is basically the same as a grand publisher conspiracy.
How silly of everyone else not to realize that dropping a client over flagged content is a deep, sinister motive, and not just, you know, standard operating procedure for every manuscript submitted today, regardless of who wrote it. Silly me, I thought reputation and business standards mattered!
P.S.
At least you wrote “Very few things are absolutely conclusive,” yet here you are trying to make what happened to Falade conclusively about racism.
I'm saying that the author could be telling the truth or the agent could be telling the truth. Who are we to decide between the two without further testing and a knowledge of AI detecting software's constraints.
"Other people haven't been caught yet" isn't a free pass to write a novel with AI. Your logic, or should I say, a lame attempt to frame Falade getting caught using AI as a racial issue, is misguided. A literary agency protecting its business standards from AI content isn't conducting a "witch hunt"; it's basic accountability. Being caught violating industry standards doesn't inevitably become a racial conspiracy just because you want it to be, just because others, with the racial profiling you'd like to see get caught, haven't been caught yet.
Speeding is still illegal, even if the cop didn't pull over every single person on the highway.
I'm stuck on this question: if a book or any text is appealing to a reader, does it matter whether ti was written with the assistance of AI? That's a different question than being dishonest about whether it was written with the assistance of AI.
I think it's inevitable that sometime soon a book will be written with AI, that fact will be disclosed upfront, and the book will be both popular and well reviewed.
To me the dishonesty is the problem, not the use of AI.
Why? If you enjoy the book what do you care who wrote it? I'm just asking this out of a curiosity because I don't buy books because of authors I buy them because of content.
I'm referring to the statement about dishonesty. We can't make deception statements about AI without being brutally honest about how publishing actually works. Conversations like this shake out this information. A published authors work is rarely untouched by many other hands, voices, POVs etc. So using deception as the argument about AI written pieces is thin.
Yes I think the dishonesty is key. And it's worth noting that big publishing houses published several openly AI co-authored books. A few got major publicity that any author would dream of. But... the sales weren't there. People don't want to buy AI work, at least if they know it is AI. That might change in the future.
I'm curious. I did a research project recently among landscapers about how they use social media to help them do their jobs. They all use YouTube videos while on the job to help them do the job right the first time. I said, "If someone asks you whether YouTube came up with the idea or whether the idea is yours, what would you say?"
They all said they would say, "I figured it out." I kind of get that because they took on the knowledge and actually did the job. Maybe this isn't exactly the same thing but it's an interesting moral analogy.
It is different but -- to me -- the shared aspect is taking credit for something that really came directly from an outside source. I think there's an aspect of human nature in play here.
If I read you right, you're saying when on a landscaping job they would look at how-to youtubes to execute a task? That doesn't seem any different from taking a guitar lesson or reading a poetry craft essay. No one expects a landscaper to come up with a new method of planting a bush or the idea of using gravel.
If you mean they copied someone else's designs entirely then I think the difference might be the type of work. A prestigious landscaper hired to design a prominent botanical gardens would probably be judged for ripping off someone else's work in a way few would judge your average landscape job. Similarly, as a writer I have a much higher--and just different--standard for myself when publishing a novel under my name than I do writing anonymous freelance work for cash. I imagine most fields are that way.
A human writing coach will not give you the text and say “now go publish this as your own.” They will more likely prompt *you* to help you generate interesting work, or describe what they are seeing in your work so you can assess how well that matches up with what you wanted to put forward. I guess one could use AI that way too, perhaps, with the right prompts, and most likely this is a better analogy to the landscaper example of using YouTube to watch someone else and then replicating those steps with the material at hand. But the minute you copy and paste and don’t disclose it, this is not “coaching” and it’s time to stop fooling yourself that this is independent artistic work.
You can tell AI not to write your story. You can use AI to re-compose a poorly written sentence. You can use AI to learn more about the location of your character for contemporary fiction, how your character might live, get around or eat. I also use Google Maps to learn more about the location. I ask AI how a specialist such as an intelligence service or police might word something. Many other details too.
I also think “self publishing” is not self publishing if you pay someone to do it for you. Same for book covers. Expensive book covers are not your own work. Make it your own by making your own. It is free.
Information is out there to be used, and we cannot keep citing the original inventor/creator of such information. Otherwise everything we write here must end with: this text is made possible by Grunt Argh, the inventor of English, by Jada Ba, the inventor of writing, by Ming Xu the inventor of paper, Christopher Sholes, the inventor of the typewriter, ……….
We should be able to own copyright on our ideas however that idea is manifested. Just because I brief eg a photographer to produce a photo to my specific brief, it should bey.copyright. he wouldn't be producing that photo of it wasnt for my brief.
Likewise AI.
It doesn't just boot itself up and produce a photo or book because it had an idea. The day it does, and that original idea is demonstrably its own, then it can be said it was its own idea.
I totally agree with you. Who gives a crap as long as the reader enjoys it? We've gotten so pompous and self-righteous like we're special artists and we have a secret that nobody knows about. Poppycock.
THIS!!! To me, this is the actual question. In simple terms, can an AI story actually be good, not in terms of prose or flawless grammar but is the story captivating. If an LLM is basically a prediction machine, can it take a story somewhere new?
I think the article covered this topic a bit. There will always be a general publishing industry that is meant for “consumption” rather than “absorption” just like there are commercial movies and Oscar worthy indie movies that will take special taste to watch. One can be a consumer and appeal to any form - actual human input or LLM driven slop.
As soon as you recognize that something was written with the assistance of an AI you get stigmatized and you lose readers. Even if you worked for months, rewritten the book 8 times, and used AI for research, analysis and proofreading only (it can’t do quality literature writing).
People will continue to be “dishonest” for as long as there is “stigma”.
Let’s fix both problems, no stigma, no dishonesty 😉
I see some commenters asking "Why does it matter if it's AI?"
If the AI authors really believe that their work is good and important, and if they have no issue with it being AI generated, then let them declare it as such. If it didn't matter then they wouldn't pretend that they created the work themselves.
I estimate that around 4% of the stories we received in our recent short story competition were created with, or with the assistance of, AI. I believe we eliminated all such stories from the running, because we were hyper-vigilant. We had a large prize fund, so we expected the scammers to come calling. Really, it wasn't very hard to spot them. There are certain tells, like an evenness to AI writing that you don't get from an authentic human writer. I guess it will get harder to pick them out in the future, which is troubling.
I think the real question here is how much work do you have to put in? Many people are spamming AI generated text with essentially no human work at all. It can be automated easily. Then on the other hand you can have heavy involvement that takes as much work as any pre-LLM method of writing. The question will be where is the line drawn, no?
AI can't creat things that are NOT ALREADY EXISTING. AI ISN'T ORIGINAL. IT STEALS FROM EXISTING WORKS TO "CREATE" A WORK. IT IS NOT ORIGINAL. IT'S STEALING.
Wrong again. I have a considerable degree of knowledge about it, having previously worked in the tech industry for 30 years. I'm also a hard-working writer and an experienced editor. So I guess you could say I cover all the bases in this discussion.
My point remains: if work has been created by AI then let the authors declare it as such. The fact that they don't tells us everything we need to know about the degree of awareness they have that they are not creating authentic material. By trying to pass it off as human-authored work, they are committing fraud, even if it's only in a moral sense.
Let them use AI; fine. But if there's no shame in it then let them be honest about it and say:
I have no problem declaring that I used AI for my book. But I didn't just write a prompt in there and say hey, write me a book about whatever this topic is. I used it to help me develop an outline and then I modify it myself and then I use it to help me do research to find things that I would have to do myself anyway but why not use a research assistant for that. For example, I could spend weeks searching Google and read it and Facebook and other areas to try to get information about a topic that I'm working on for my book when an llm can handle that in a matter of a minute or two. I write the paragraphs. It's my idea and my thought process. And I let AI help me with making sure that I'm consistent across the board. For example, I loaded a couple other books I wrote into ai and I had it developed what's called a DNA guide which standardizes my sentence structure and my word structure and grammar and other areas that are unique to me. And then I use that to test each chapter to make sure that it follows those standards. What's the difference between doing this and sending it out to an editor and letting an editor go through and do the same thing? In summary, I don't think that it's right to allow the AI to generate the entire book from a simple prompt of write me a book about something, but I don't see any issues with using it in order to assist with research, standardizing grammar and sentence structure and the rest as long as it's still my voice.
I have no real problem with that use of it, either. For me, the issue is when 'creators' use AI to generate the paragraphs of a story/novel, or the stanzas of a poem. Research is perfectly legit with AI, it's great at that stuff, though does tend to throw in incorrect information from time to time, so verifying via other sources is probably a good idea. Dictation tools like Wispr Flow use AI; again, it's perfectly legit in my opinion to use such tools.
Just don't (and I'm not talking about you here, just in very general terms) get it to write a story/novel/whatever and then tell me that [insert name] wrote it themselves, and then call that art.
Or do, because it's a free-ish world.
To be honest, my real issue is when-AI generated material ends up in our submissions queue. Genuine artists are vying for limited publication opportunities for their work, and I feel there's something immoral about asking an LLM to create it and then claim it as one's own. It's like me buying a painting then telling my friends that I painted it.
I agree with you 100%. I'm a data scientist and I do a lot of work in AI engineering and agentic development. I use AI extensively in research and statistical modeling and even report writing. But in every single case, I am the human in the loop. Nothing gets posted or released until I have had a chance to review it personally in the same goes for my book writing. I wrote a book called "service dogs for mental health" and it is done okay and I used AI for a good part of the research. For example, when I had to really get into the Weeds on what the Americans with disability act said, particularly titles two and three, it was easier to use AI to generate the language and the citations, but I had to go back in and check those myself to make sure they are accurate. In this newest book that I just wrote, there's 101 citations and I committed to the publisher that I would attest to having reviewed every single one of those personally, meaning pulling the book or the article or the publication or whatever it was and verifying the context, the content, the page number and everything else related to the citation. I just think that there's this blanket fear of what AI is going to do to the artistry if you will of authorship and I think it'll become quite apparent the difference between those that have a powerful human in the loop model compared to those that rely on autonomous writing.
I'll add, Frank, that I use AI for all sorts of things, from recipe recommendations to understanding the rules for the calculation of my golf handicap, to creating VBA macros for Excel. I don't believe it's a bad thing in and of itself, in other words.
What strikes me in all of this is how much the industry is still operating inside a causal worldview — where authorship is treated like a chain of custody, creativity is imagined as a single point of origin, and the fear is always about what a decision might cause downstream.
But for the reader, none of that matters.
The “man behind the curtain” isn’t the site of value.
The value happens entirely in the consciousness of the receiver.
Every creative act — painting, filmmaking, symphonic composition, novel‑writing — has always been collaborative, derivative, and emergent. Da Vinci had apprentices. Beethoven needed an orchestra. Spielberg works with hundreds of artists. A play isn’t the playwright; it’s the performance.
The work is never the artifact.
The work is the experience the artifact generates.
That’s why the AI panic feels misplaced. It’s trying to force a square peg (a distributed creative reality) into a round hole (a causal model of authorship). It’s the same chicken‑and‑egg problem: a paradox created only because we insist on linearity where none exists.
What’s really happening in publishing right now isn’t “AI disruption.”
It’s transition dynamics — the field tuning itself in real time as old assumptions about creativity collide with a new reality.
And the only thing that ultimately matters is the coherence of the field between creator, work, and reader. Everything else is noise.
While you can take this stance, it is obviously false to claim this is the universal opinion of readers. Many readers do care about questions of authorship, plagiarism, falsification in memoirs, etc. We wouldn't have these scandals otherwise!
So many contributions to the world happen from standing on the shoulders of others. But in our individualistic culture we're encouraged to take full credit for the latest creation without acknowledging the work of those before us that allowed us to progress our ideas. Maybe big stories like these will move us towards full disclosure and acknowledgement, which might not be a bad thing.
In terms of using AI for research I don't see how this is any different to using other digital tools for research (google, bing). Of course writers are using AI for research.
We all stand on the shoulders of others, it is true and should always be acknowledged. I think this applies to knowledge of the human kind and be-ing of the existence kind as well. This AI morass is opening so many lines of enquiry and confusion it seems we will have to work it out in the commons, together, writer to writer to reader to thinker…..etc. I often think about the uniqueness of creativity and the direct outcomes of coming into contact with ‘real’ creativity. What is real creativity anyway? It feels as though we are treading water on a long line of ancestral knowing and in many ways rehashing the rediscovered and the previously known. One thing is for sure: the days I feel genuinely inspired by what I think is unique creativity is always brought to ground by the fact that somewhere someone has had the same thought at some point. Makes me feel like a pop-up consciousness @Jordan Clist.
As a counter example, my kid was a very avid Harry Potter fan. After the writer started taking certain positions things changed, especially when those positions were used in attack mode. They will not do anything that enriches her.
It is not solely on the text that enjoyment and interest lies. Readers are a (diverse) community.
I can appreciate that. But how much of that is learned behavior? Or peer pressure? We enrich all kinds of people whose views don’t align with our own. But I applaud the integrity.
I'm not sure if I understand your comment. They are not enriching the author of the Harry Potter series. They will buy nothing connected to it. No streaming, replays, re-releases, not a penny directly or indirectly. [And neither will I].
This set of dictums is riddled with holes like "A play isn’t the playwright; it’s the performance." That's illogical, because whoever claims that a play is the playwright? A play is the creative product of a playwright and exists whether it's performed or not. It is an artifact. The work is ipso facto an artifact.
Of course the artifact exists. It’s like the idea that, “If a tree falls in the woods with no one around does it make a sound.” If a playwright is just writing the play for the sake of writing the play, it might as well be a diary entry. If the play isn’t to be read or performed for an audience what purpose does it serve? The communication process (of which all artistic expression is at bottom) is incomplete without the sender, the message, the medium, and the receiver.
How much do you think has been written from 1900 to 2026, including articles, books, magazines, journals, plays, films, blogs, and now podcasts? Now, out of all that, how many sentences do you think exist in all that writing? Well, it is somewhere between 50 and 100 trillion sentences (that is only the last 126 years). Meaning, depending on what you are writing about and the style you are writing in, 10 to about 30% of your sentences have been written before. There are only so many ways to describe the everyday experiences or the varying structures within maths. Computational: any given language has what seems like endless variations of itself; the problem is the human experience does not.
The perspective or perception (the intention to preserve an idea or an artifact as a witness) turns something that is yours into something original for you, and it could make someone else think or see in another way.
Using ai and relying on ai are the same thing imo. If at any point in the writing process you feel you absolutely must use ai for anything, even if it's just for inspiration, instead of just doing it the old fashioned way, you're relying on it for a part of your writing to some extent. You're the one writing the prompt, yes, but you're still depending on what it answers back to you to help make your final decisions for the final piece.
i still don’t understand your point. his agents were the ones who brokered the $2 million deal. the deal didn’t happen out of thin air. the agents planned the auction, they invested a lot of work in making the book deal happen. that’s what agents do. they make the best deal for their clients — and they work on commission. a $2 million dollar book deal means $300,000 commission for themselves.
i still don’t understand your point. his agents were the ones who brokered the $2 million deal. the deal didn’t happen out of thin air. the agents planned the auction, they invested a lot of work in making the book deal happen. that’s what agents do. they make the best deal for their clients — and they work on commission. a $2 million dollar book deal means $300,000 commission for themselves. that’s an enormous pay day.
they loved the book and they believed in the book. and they worked hard to orchestrate an amazing book deal of $2 million dollars.
they knew he was black when they orchestrated the book auction. the publishers didn’t pull out of the deal. the agents themselves pulled out of the deal after they became aware of issues involving AI use. they had previously vetted the book with regard to use of AI, and initially they were satisfied. then after the book deal, they learned some new information, which made them cancel the deal. they are not disclosing what the issues were. so no one is in position to argue about what they learned — we simply don’t know what they learned that made them withdraw their support.
we can only assume that what they learned was something substantive, because it caused them to sacrifice their $300,000 commission.
he has retained an attorney and he plans yo sue. if he sues, then all the details will get hashed out in the trial. it doesn’t make sense to jump to conclusions without knowing the details. but right now the only ones who know the details are the author and the agents. the publisher with the winning bid doesn’t even know the details.
the larger point is that we are in a very choppy period for publishing. AI has upended all the rules, norms and expectations that have been in place for centuries. there is no reliable way to determine AI use, except by the author’s acknowledgment and admission. so in this case, it is very hard to imagine what the agents learned that caused them to withdraw their support for the book. the writer is not admitting to AI use and he has hired a lawyer. so it this story is not over yet.
the thing is, i doubt any of these books are gonna be fun, considering all the repetitive, predictable and boring stuff ai writes. i think allowing this kind of stuff in the market may result in a massive amount of crap that keeps growing, possibly overshadowing real, good stories from really talented authors whose life are built around their writing
But that's simply not true. Depending on how the parameters are structured, how the prompt is worded, how the contextualized engineering is done, the temperature, the p-value, all the other attributes and parameters and components associated to it produce completely different outputs.
I didn't read the article because I've heard a lot lately here on substack that the tools that are coming out to verify AI use or not in articles is having a lot of false positives. That might not be the case here because you would think if the agents pulled it and that costs them a lot of money there might be more to it. But relying on a computer program to verify if something else is a AI is a little crazy and symbolizes all that wrong with rapid AI adaptation.
I write for a living where sometimes AI is accepted and sometimes it isn't. I'll say this: I've been flagged for AI when I didn't use it and have skated past these flagging systems when I did. Using technology to verify technology use is not at all helpful or accurate.
Since the beginning of time, all ideas — written or spoken — are derived or borrowed from somewhere. Using AI to help develop a story is no different than using a thesaurus to discover different words. If we’re accepting AI chat bots as necessary in society, then we need to accept that AI will help write stories, ideas, history, etc… from now on.
You’re right, we create differently from what we’re exposed to. Ideas from other sources are heard, then reacted to. Nothing is original. Everything in existence is created from something else.
I don't see the issue with using AI for research and copy editing. My novels are set in 1993 and in order to accurately describe settings and scenes, LLMs are extremely helpful in jogging my memory.
Writing was once controlled by the few who possessed the education, time, and institutional access to produce it. Then came the printing press, public education, the typewriter, desktop publishing, the internet, and self-publishing. Each innovation broke a form of scarcity, threatened established gatekeepers, and inspired warnings that inferior work would flood the culture. Some of today’s resistance to AI belongs to that same pattern. But AI also introduces something genuinely new: the printing press democratized the distribution of human writing; AI democratizes the production of plausible language itself. The real question may not be whether AI writing is legitimate, but who gains the power the old gatekeepers lose and who ultimately owns the new “printing press.” The more useful historical lesson is that democratization and exploitation can occur simultaneously. The printing press liberated knowledge, but printers became new gatekeepers. The internet allowed anyone to publish, but platforms eventually controlled visibility. AI may enable people who were excluded by education, disability, language, money, or professional networks to express complex ideas but a handful of corporations may own the systems through which they do it.
You nailed it. The human owns the complex, non-trivial themes and ideas. They now have AI to help express themselves. They should proudly declare that they used AI to create a manuscript that would otherwise not exist. Then it's for the readers to judge the value of the content.
Truthfully, his has become a witch hunt. I'm not defending AI, God, no. But the fact that we have to dumb down our creations to not be able to be "detected" as AI is insanity. Just bcs someone wrote something good, doesn't mean they used AI. Just because someone can write fast, doesn't mean they used AI. They were robbed for something alleged that wasn't even proven. Ridiculous.
Maybe I'll vex these witch hunters by proposing this:
I've decided not to use AI.
Instead I will move to a country where they don't have any child labour laws and teach kids under the age of 5 to read all the great works of fiction and non-fiction. And then - paying them nothing but scraps and sesame seeds, I'll get them to write my next big book to be sold for $200 million (upfront of course) and no one will be the wiser. No one will even suggest to visit me at my hideaway in international law grey zone or look at my compound to see these children writing and typing on refurbished MacBooks and Surface tablets. They won't even bother with sitting with me and just talking about my work to see if I actually have the intelligence to create such a work. They just pay me... and then ... later... discover I was using kids... illegally... to write my $200million manuscript.
This AI witch-hunt has become its own sub-genre in the long line of people who are unhappy their mediocre writers and think other mediocre writers are achieving great results by simply typing a prompt "make me an award winning novel, chop chop."
In truth, they are envious they didn't think of it sooner. But now... they're unhappy it's not actually possible since they'll be discovered at some point. So really... this whole accusation is coming from people who can't see themselves writing so brilliantly they'll be the next [insert writing hero] unless they got the boosted abilities afforded by some AIs. BUT... they do not want to rely on AI. Circular argument for people unhappy with the challenges of real writing and creative output.
Has Proust been run through Panagram? Hemingway? Dostoyevsky? Have any of you written a book with the help of AI? How different it is from having an editor? or a white board? Do we have to scale back to chisels and stone to be authentic? Do only those who can afford a human editor or have the right connections or pedigree merit access to formal publishing? How much did the editor, the sales team, the publisher, the designer, the cover art affect the final output? How many drafts were there? Why is that acceptable? Shouldn't the prose be pure and untouched? Was research done? Did the author look at a royalty free picture of a castle to help describe their castle? Did they listen to music to generate a vibe while recording notes? Should Taylor Swift be collecting royalties?!
“I also would not hire a ghostwriter to write my work. I became a writer because I like writing. The appeal is in the act itself.”
My favourite few sentences. It’s why I read. It’s why I write. It’s why I sell books. It’s the appeal of the written word, the gift of an author/writer to me.
My writing may never be published, it may always remain a “draft”, but it is all mine. I try to imagine Austen or Tolkien using AI … yeah, nah. AI may have its uses, it may be a tool for checking etc. what it isn’t is living and breathing and experiencing life. What it isn’t is human.
Quite a few comments here are hedging by pretending it isn't possible for an LLM to generate text without lots of human work. This is simply not the case. You can go to ChatGPT right now and ask it to "generate a prompt for a poem" and then use that prompt to generate the poem. If you believe that still makes the prompter an artist, well, I respect the intellectual consistency I suppose! But it is dodging the real question to pretend the only LLM art out there involves tons of human effort on behalf of the prompter.
Here's an AI author who says the pump out a novel in 45 minutes and claims to earn a lot of money from it: https://www.nytimes.com/2026/02/08/business/ai-claude-romance-books.html
Good for them. Everyone should be that efficient.
Anyone can be! Hell, I just completed a marathon in under half an hour! Just me (plus one tool [my car])
If that's what they think creating is, i don't need to buy their work. I can just make it myself.
Exactly. It is without value. No imagination or experience other than what has been stolen.
A poem is one thing, a whole book is an entirely different beast. Go to ChatGPT now and ask it to write one chapter of a book, sit and laugh. It can’t. It will give you 5k words but 70% if not more will be one-liners.
LLMs cannot write books. They cannot write even one good chapter of a book. It doesn’t have the computational power to do so or the experience to weave through the intricacies of books. Please try it yourself and let us know
If you agree LLMs can already write poems and short essays, I think it's obvious their ability to write at length will increase as time goes on. We've seen this in video outputs, which use to fail at keeping objects consistent for a few seconds and now can do so for longer.
So, I think we still have to wrestle with this basic fact that LLMs allow a full range of human collaboration including 0%.
I don’t think they can write poems either. And they cannot write short stories. AIs will get better I’m sure, I’ve noticed the progress, but as long as it generates rather than imagine, it can’t complete a story.
Well they obvious *can* write poems and stories. Whether they are *good* or not is another question. I agree that LLMs are not likely to create great or innovative literature by themselves any time soon. But most of what people read and buy is formulaic and I could easily see LLMs competing there. Really, they already are. TiKTok and Instagram Reels are filled with AI slop racking up many views and I bet AI generated clickbait websites get decent traffic. A recent study showed AI books getting financial returns in the self-pub Amazon marketplace.
There are already apps with AI stories and millions of downloads. Many (All?) of them are like a series with neverending episodes. And they cost an absolute fortune, you could end up spending 100s to hear the end of a story, a book can be bought for 10-20 or even a lot less. Sad.
We can’t really say that whether they’re good or not is another question. Everyone can write after all and only major writers get that immortal fame.
It doesn’t also matter whether AI generated books bring money. With today’s marketing techniques people are buying virtual things.
We’re talking here about books that matter, with depth, innovative and/or masterful literary techniques. AI is far behind in terms of creating such books, and believe me, I’m a big fan of AI. As I mentioned I do use it and have put together a process that works and helps me advance faster. But left to itself it cannot, as an LLM, create anything that comes close to even an average human writer. Maybe the next General AI would. I’ll try it and will discover. But today, LLMs assemble, they don’t create.
You must not be much of a numbers man.
The specific numbers that are worth talking about are the 60/40 mentioned in the article. For people that use AI responsibly, that's not a bad ratio.
The more obvious point: bidding war and many professionals somehow overlooking an AI contribution? If you're so academically inclined, Mr. Intellect, then why do you point the finger at a machine rather than the people disregarding it?
I'll bet 100,000k that all of them knew but the general swing of culture right now is AI = replacement = evil so they pulled their investment proposals to cover their asses.
Let's be real real here. You can romanticize anything you want, I certainly do. But don't sit on your soap box with your romanticized rhetoric. The publishing industry has been shitting on and eating the souls of anyone that's not jk rowling. And now you're protecting them because there's a new "evil" in town?
Get off that soap box and put it in your mouth.
I'm replying to my own comment because I don't actually want you to go through soap in your mouth. It's not pleasant. But read my comment as something that might be thought provoking. Instead of the human provoking I did to you.
I've tried this as an experiment on a short story. To generate anything without obvious tics and beyond the lowest literary quality is death by 10,000 prompts. It is easier, at least in my case, to in fact write the thing I want generated myself. The only utility I've found in it is in some fact checking and research, in which case it is glorified Google. Caveats apply there, too.
It's way easier to write it yourself, and limit the role of the AI to analysis, critique, proofreading and line editing. With that, I was able to finish a book in 4 months when it used to take me a year.
I'm curious what LLMs are speeding up here. Maybe my process is just very different because I don't think that would speed things up at all for me. Proofreading is a pain but certainly not 8 months time. By "line editing" do you mean you let the LLM rewrite your text beyond mere proofreading?
Sorry I missed this. I'm happy to share with you what I call the ACE method (Analysis, Critique, and Editing). I don't know your personal process; mine is to write, rewrite, and keep rewriting until I think it's done. That's usually around version 7 or 8. Through this AI-assisted technique, I'm down to 4 rewrites.
I write a chapter, run it for analysis, and the AI instantly gives me twelve pages of analysis to ponder. Not to accept, but to think about. Then it critiques my work: what's good to keep and where I'm short on details, where I repeat myself unnecessarily, or where I expand too much on an idea. Then I rewrite the chapter and run it through A and C again. After a few more proposed changes, whether I accept them or not, I ask for editing and move to the next chapter. Once I finish the latest version, I start over.
What you get with AI is immediate feedback; some you disregard, others you use to better your work. At the same time, each version narrows the AI's thinking. Left without the narrowing, it will roam freely in the billions of oceans of information at its disposal, uttering unrelated and sometimes ridiculous notes. Boxing the AI in the book's world led to higher-quality analysis, critique, and editing. It becomes a reader and an editor working beside you at all times, giving you immediate feedback to use or disregard.
It's a fun process, and the fast pace is motivating.
Before, I needed to rely heavily on friends' input or pay a professional - and wait for their input. Today, I'm sending my final version to a professional editor, and it's returning a week or so later with far fewer notes and proposed changes.
I sincerely invite you and other writers to try this method. You write, you decide, you're in control; only there's someone next to you assisting at all times.
Proofreading, analysis, critique??? Wow, what a waste of time. And thanks for obliterating thousands upon thousands of jobs.
Why don’t you write your comment with a pen on paper and send it by courrier to Substack if you’re so worried about saving jobs. Self-righteous much?
Dont worry, youre blocked and I will never read or buy anything you write, editing is one of the most important parts of writing a book, destroying the planet for your convenience is the ultimate in entitlement and indicates that you are not a very good writer.
Twatwaffle 😂
Theres a difference between quality content ai can generate but its text limited so youd need to piece together
But its memory starts to forget ai is a tool afterall
Considering how many ghostwriters are responsible for NYT Bestsellers and works that make publishing houses $$$$$$$, I'm struggling with the AI debates in this arena. Even high quality AI pieces require extensive skills when it comes to crafting prompts, revising, editing, etc.
*
Edit: Reading through the responses has reinforced that this conversation extends far beyond AI. We've touched on ghostwriters, copyright law, plagiarism, originality, intellectual property, publishing, ownership, and whether there is even such a thing as an original idea. These aren't separate conversations, but expressions of how we've commercialized human creativity.
Human beings are generative by nature. Every company, invention, book, song, painting, technology, and economic system began as someone's capacity to create. Creativity is how we translate our time, attention, experience, relationships, and sweat equity into material value.
Scarcity consciousness (there's not enough for me) shifts our identity away from our capacity to generate value and onto what we've already produced. Once that happens, our worth becomes increasingly tied to external validation of those creations. Publishing deals, bestseller lists, copyright, ownership, attribution, awards, recognition, market value. They stop functioning as ways of governing the commercialization of creative work and become measures of personal worth, turning the artifact into a proxy for identity.
That's why these conversations remain so emotionally charged. A creative work can absolutely be copied, reproduced, commercialized, infringed upon, or stolen. The human capacity that produced it cannot. When we lose sight of that distinction, we begin protecting artifacts as though they contain the source of our value instead of recognizing them as expressions of a capacity that remains inherently human and endlessly generative.
I think in many ways comparisons to ghostwriting is fair, but also... using a ghostwriter is not really artistically respected. It's mainly used by businessmen and celebrities who can't write or commercial fiction packaging companies that often release books under made up names to control the profits.
Yes. Publishing houses had to pivot years ago to start purchasing attention, by way of giving book deals to people based on social media status and reach - many of which require ghost writers.
Then there is gamification of the NYT bestseller list, which is why many titles have a dagger next to them. James Clear dominated with Atomic Habits because allegedly he and his team figured out the precise amount of books that needed to be purchased weekly to maintain his position. Not because it was the best book on the market.
All this to say, crratives need to stop measuring their success/ worth or placing value in these systems and structures because they are inherently untrustworthy.
What authors should aim for: consistent sales over time. That’ll get them much farther than, say, a week or two on a bestseller list & then a drop in sales. It’s also widely known in the industry that the NYT list is curated. That doesn’t necessarily mean the “best” books get on the list. Merit really has less & less to do with being a bestseller. It’s more about what consumers are buying.
100% this. I was just at a dinner where they were discussing plans to game a book onto the NYT bestseller.
A friend of mine was absolutely gutted after their first book went #1 organically and the second got a dagger upon release because their publishing house didn't set up different purchasing portals for bulk pre-sales. I didn't know anything about this system until I started seeing what gets sent to the authors from NYT.
What I’ll say about that is that your friend would still receive the benefits of having a bestseller: store placement, consumer recognition. Most consumers have no idea what that dagger means. They just know the book is on the bestseller list.
and ghostwriters stand in line to be selected..... These writers make oodles of $$ because high-profile books sell and many of them sell quite well....
After stories I've heard first hand about the process of becoming a best seller....never in a million years would I sign those contracts. Literary agents, editors, publishing houses...so predatory. And it works because so many people are seeking validation outside of themselves to feel worthy of their creations.
you make an excellent point….historically, at least since legacy publishers came on the scene at the turn of the 20th century, the ‘gold standard’ was acceptance into this club….we are in a seismic shift, undergoing a radical change that I think those of us in this world lose sight of….I’m interviewing @Kerri Schlottman soon on my podcast…a bestselling author, #literary, who shares her road to publication stories…it’s not easy to figure out and she openingly talked about this at a conference where she keynoted…can’t wait to continue our convo..and will post here…
I will keep an eye out for this!
Your point about ghost writing is on point. What difference does it make how the book was created as long as it's good. The amount of research, input, revising, and editing required to produce something of quality with AI is extensive. Most people who don't use it don't realize that. Also, I put a college paper I wrote 10 years ago into an AI detector, and it said my paper was 78% AI. AI writing had progressed to a point where algorithms think human writing is AI. They are completely unreliable.
Once again, that's the thing. If so much work goes into it regardless, WHY do people who use AI to write their books feel the need to use it? Why outsource their thinking and researching that why? Why not just do all those tasks they're assigning to AI themselves if it's gonna take them sooooo much work and skill input either way?
Agree, the argument is mute. Can one say that cavemen art is the only true art because it was solely done by hand? That a more and more sophisticated brush didn’t make painting more precise, more detailed. I’d like to see anyone here looking at Picasso in the eyes and calling him a fraud, or removing his work from auctions… it’s a shame they pulled a fantastic work because it was made with the help of the newest tool. It’s also a shame that the writer is pulling the race card - but that’s a different subject.
Keep writing and use what makes your writing better without shame or fear. In a few years all this will be in the past and people will be arguing about another silly subject.
The problem with this argument is that it LLMs are a "tool" that can do .0001% of the work or 100% of the work, plus any % in between. This is what makes conversations so tough. Some uses of LLMs are like a paintbrush. Others are like hiring a ghostwriter you collaborate with. And others are basically plagiarism (passing off someone else's work as your own) or fraud.
Picasso used paintbrushes, but without Picasso the brushes and paint would do nothing. OTOH, I can go to ChatGPT right now and say "write a poem" and it will write a poem with zero input, ideas, technique, or anything else coming from me. I just did that now and here is the response
"Where the Light Goes
Morning slips softly through the trees,
painting gold on every leaf.
The wind carries yesterday away
like a letter no one needs to keep.
A river holds the quiet sky,
clouds drifting where they please,
and somewhere in the distance
a bird sings through the breeze.
We spend our lives chasing moments,
afraid that they will disappear—
but some things stay long after,
like sunlight lingering here.
So let the day unfold slowly.
Let your heart be where it goes.
Even when the road is hidden,
there is light beneath your toes."
Would you say this poem is mine? That I should get any credit for it when my entire prompt was "write a poem"? I think 99.999% of people would say no. This is an LLM written poem. The human had no notable involvement. OTOH, if I wrote a poem myself and then asked ChatGPT to offer feedback and it suggested tweaking a line or two, most people would say I wrote that poem.
So the whole issue with LLMs in art comes down to deciding where the line is. When does it go from tool to collaborate to some form of fraud or plagiarism. And yes, it is plagiarism to claim work as yours which you didn't do. And this is not hypothetical. Lots of people are doing more or less what I just did with the above poem and then submitting them to lit mags or uploading books of them to the Kindle Store.
I'm sorry to say this isn't a poem, and if you think it is, I understand why you're worried about AI taking your job.
The argument is "mute"? Do you mean "moot"?
You're all over this comment section claiming that people shouldn't care about AI use in fiction and predicting that, someday soon, nobody will care.
People care. Especially when it comes to fiction. The majority of readers still want to know that the prose they are reading was produced by a human being. These controversies wouldn't occur otherwise.
AI is taking over the creative arts with the same inexorable incrementalism we saw when we gave away our privacy in the name of convenience and safety, but that doesn't mean writers should be uniformly grateful to the tech companies.
We are already seeing excessive reliance on AI in classrooms. Offloading cognitive tasks erodes core mental skills like critical thinking and memory retention. I don't even need to explain why these things are valuable to a writer.
You've claimed here that an LLM can't write a short story. Have you read the short story that won this year's Commonwealth Prize? I'm not talking about copying and pasting it into Pangram. Just read the first few paragraphs. It's incomprehensible. Laughably bad. Yet it hoodwinked the judges.
I don't mean to be so hostile but it's exhausting to see resistance to AI characterized as the ravings of backwards Luddites. There's a bizarre cult-like fervor to some writers' embrace of AI. Any writer who can't see the obvious value of the iterative drafting process that literally IS writing should be demoted to prompter.
Sorry for the typo, I should have ran my reply through ChatGPT 😀
I didn’t say people shouldn’t care, all I’m saying throughout is that LLMs cannot write one chapter of a book and even less an entire literary work. Therefore there’s no need to argue right now. Later, maybe, but right now a writer’s role is very safe.
I can’t speak for judges of literary work who got fooled, their problem, their reputation as judges.
Criticizing the use of AI in education is also moot, the same was said about the internet, remember? How student copy-pasted from Wikipedia or other published works. Still happening among universities’ PhDs btw.
You need mental skills for them to erode in the first place. A degree does not give you that. Today it’s just three words on a resume. Education and Academia, as presented, have failed to produce an awoken class and are now corporative in nature, spreading toxic philosophies and culture wars that challenge critical thinking and are rewriting what memories should retain…
I’m not saying you’re backward, I don’t even think you’re hostile, you’re writing your opinions elegantly and you have all the right to do so.
As I do. I have the right to adopt a new technology that helps me write faster, and better without this “shaming” without thinking. I’m forcing no one to use it, I found a way, using my analytical skills, developing communication methods with a new non-human intelligence to help me write books faster and better.
Many inventions before did the exact same thing (internet, MS Word, a typewriter, the printing press, all the way to the invention of pens and paper) each of these inventions made our work better and faster yet no one argues about them anymore. Technologies have helped humanity, that much is undeniable. And every new technology brought with it a wide range of arguments and opinions, until time proves their worth and their shortcomings. LLMs are the same. In writing, in art in general, they remain tools rather than agents and require the artists’ full involvement, all they do is save time. Kind of like the bean can you simply buy at the supermarket. Does a farmer still do it by hand, caring for every plant, spending days collecting, gathering, preserving, bagging, transporting, selling… should we shame modern beans producers for using all the technologies available? No one is. Why shaming writers who use new tech? Why was a great book canceled knowing too well that the writer did the writers’ fundamental work?
How do you know it's a “great book”? Nobody has read the manuscript. It's not even available online. Are you just assuming it must have been great because of the advance?
The author in question hasn't been able to provide much in the way of working documents, which is odd. Writers love to talk about process. (Probably because nobody ever asks us.) Moreover, no literary agent would throw away an advance that big without good reason.
This “it's a tool, just like any other” refrain is either a canned speech or willful blindness. For all of human history, tools have been passive extensions of the human body, like a hammer (hand/arm) or a camera (eye). AI actively generates content, makes decisions, predicts outcomes and allows humans to offload cognitive tasks. It's much closer to an autonomous agent or digital worker. It can used as a tool but it's not just a tool and repeating that same line over and over won't make it so.
Writing is more than merely generating prose. It's a process. A valuable one. Outsourcing that process because you're busy and it's faster misses the point, IMO. Writing is about quality, not quantity. Churning out books faster does not mean they're better.
Yep, you're free to use AI. Just like I'm free to think it sucks and writers who adopt it wholesale are deliberately selling themselves and their writing short. But demanding that everyone else get on board while making the same tired points about it's “inevitability” is like throwing your lot in with Goliath and wondering why David doesn't like you anymore.
You don't get to use a technology that does research, edits, and writes for you AND demand respect for that choice.
A book that gets a $2m contract from publishers is usually a very good book. And we've heard your refrain over and over whenever a new tech takes its first step. All your purity arguments fall apart each time, and the technology wins the day. Then you complain about being left behind.
You're putting words in my mouth that I never used, exactly like someone who can't hold a serious argument. I don't care about your respect, and even less your opinion once you start playing these losers' tricks. Go bore someone else please.
That is wild! 78% off target? Is junk!
From what I understand, via research, is that you cannot copyright AI content. All AI content falls in the public domain. This means if a publisher invests a lot of money into a book, once it is published anyone can copy it and sell it. Without a copyright, the publisher can’t do anything about it. Who is going to risk two million dollars or whatever when this is the case? If you study copyright laws as of today, you can copyright the human parts but not the AI generated parts. So, even if two percent of the book is AI, you can’t copyright it as a whole.
How in the world can anyone ever know whether it was AI or not? Do you know that a very well written piece using proper grammar and good sentence structure often comes up as looking like it was written by AI. It's time to get over it and on with it because it's here to stay.
According to the copyright office, they can. I only looked into this as I use AI to assist me and wanted to be clear as to what I can and cannot do should I ever want to publish anything. Personally, I could care less if people use it or not. Actually, I am glad some people do as there are people with brilliant ideas who cannot express them properly in writing.
You should look at the actual law. Material created with AI support is copywritable unless something has changed in the past few weeks. AI is presently seen as hired help. Many authors employ teams of writers already and have been for decades.
I guess the debate is “support vs generation”. I just looked at it again:
Based on an analysis of copyright law and policy, informed by the many thoughtful
comments in response to our NOI, the Office makes the following conclusions and
recommendations:
• Questions of copyrightability and AI can be resolved pursuant to existing law,
without the need for legislative change.
• The use of AI tools to assist rather than stand in for human creativity does not affect
the availability of copyright protection for the output.
• Copyright protects the original expression in a work created by a human author,
even if the work also includes AI-generated material.
• Copyright does not extend to purely AI-generated material, or material where there
is insufficient human control over the expressive elements.
• Whether human contributions to AI-generated outputs are sufficient to constitute
authorship must be analyzed on a case-by-case basis.
• Based on the functioning of current generally available technology, prompts do not
alone provide sufficient control.
• Human authors are entitled to copyright in their works of authorship that are
perceptible in AI-generated outputs, as well as the creative selection, coordination,
or arrangement of material in the outputs, or creative modifications of the outputs.
• The case has not been made for additional copyright or sui generis protection for AI-
generated content.
The Office will continue to monitor technological and legal developments to determine
whether any of these conclusions should be revisited. It will also provide ongoing assistance to
the public, including through additional registration guidance and an update to the
Compendium of U.S. Copyright Office Practices.
Thanks. I have dug a little deeper to understand what constitutes material contribution. Plot development, character definition, world-building, all work. And in fact AI support for world-building is very helpful.
"AI support for world building" is actually scraping of worlds built by other HUMANS!
No, hardly. The value of AI in world-building lies in consistency checking, inconsistency discovery, alternatives evaluated against proposed new mechanics.
I suspect you think AI writes novels, I disagree. AI, by itself, does absolutely nothing. Someday, we may be able to task AI with creating a book with the prompt “write a good book”. That would not work today. But if it does materialize, when it does, resistance will be futile - but there are other responses.
I think you're assuming AI is only functioning like knock-off factories. It's not. Human also absorb ideas and concepts their entire life and create from experience. This is why I'm fascinated by the human obsession with ownership and control - especially in creative fields. We never truly own anything. We never truly create anything from scratch, but our egos are developed through serious lack and scarcity consciousness, which perpetuates a deep desire to death grip what we 'own', instead of embody the essence of how we created in the first place.
What makes AI dangerous is it is entirely a product of and servant of corporate capitalism designed for generating corporate profits. Its incentive structures and whatever legal and political structures develop around it will be designed to protect its value to privately owned corporations at whatever cost and future blight it has for the great majority of human beings and for human culture into the future. Whatever tiny shoots of human artistry, community or creativity may appear will be crushed or turned against humanity by the overwhelming and deadening force of corporate capitalism. Look around if you think that's not true.
That and the fact that it appears to reinforce a false view of the self as a sovereign entity. It is a form of power. As such it will be unevenly distributed. That uneven distribution will further the divide between the ultra-rich and every one else. It is the ultimate extraction technology.
What should be the proudest moment in human history, will actually be the worshipping of an AI idol, and some severe corruption due to the further increase in power inequality.
That is the real problem, but there is hope. Visit my page and read “Is the Internet Eating Itself.” Warning, you have to have an open mind in relation to astrology
The phrase “There is nothing new under the Sun” comes to mind.
What is ever created from scratch??
You're now wading into a territory I'm very interested in, which is the human obsession with ownership/control.
Yes, this opens up huge cans of worms.
There are already several copyright infringement lawsuits in progress from several content providers who are suing AI companies for stealing their content.
A LOT of AI content was already stolen and violated copyright of the source providers. It is well known and documented that AI companies scrape content without regard for copyrighting.
The plot thickens.
COPYRIGHT, AFTER THE HEIST
You have robbed us.
Not metaphorically. Not poetically. Not in some abstract, academic sense that can be debated in conferences and footnoted into irrelevance.
You have robbed us—clean, deliberate, and systemic.
Our minerals.
Our natural resources.
Our history.
Our heritage.
Our arts.
Our craft.
Our stories.
Our religions.
Our legends.
Our myths.
Our tales.
Our words.
You took what could be carried. You extracted what could be sold. You catalogued what could be studied. You renamed what could be owned.
And what you could not absorb—you vilified.
You called it primitive. Pagan. Backward. Dangerous.
You erased it from classrooms, mocked it in pulpits, outlawed it in policy.
Then you wrote books about it.
Then you copyrighted those books.
Now, after the heist, after the archive, after the rebranding—you turn around and lecture me about ownership.
Copyright.
You say it like it is sacred. Like it emerged from pure intention. Like it was not built on top of everything you already took.
And now, we arrive at your newest anxiety: artificial intelligence.
You are outraged that it has been trained on your data. Your books. Your images. Your systems.
But let’s be honest about the lineage of that data.
What exactly is “yours”?
The stories shaped by structures you inherited from us?
The aesthetics influenced by forms you extracted from us?
The knowledge systems you once dismissed, now repackaged and monetized?
You are worried about theft now?
Interesting timing.
Because from where I stand, AI is not the beginning of the problem. It is the exposure of it.
It is a mirror held up to centuries of accumulation without accountability.
So no—I will not perform outrage on your behalf.
I will use it.
Unashamedly.
I will use it because my fingerprints are already in the foundation of what you call “your” data. Because the sweat of my ancestors—and my own—has already been absorbed into the systems you defend.
And if I cannot use these tools to build something entirely new, then I will use them to dilute what you stole.
To remix it.
To fracture it.
To reclaim space inside it.
You call that misuse.
I call it correction.
You want respect for copyright?
Start by acknowledging the original theft.
Until then, don’t confuse your laws with justice.
Named it.
Authors have already been doing that for a very long time. Take John Carter to Flash Gordon, to Buck Rogers, to Adam Strange, are they not all basically the same (different names, different environments - if there are only 7 to maybe 36 fundamental main story themes that can exist, why question originality now)? There is a confusion in comparing one timeline of literature to the current timeline of literature. If, say, Star Wars and or Harry Potter were released now, instead of then, would they be in dispute (potentially even algorithmic flags for plagiarism)? Is not every story built on top of every other story? What is then being potentially judged is an individual's knowledge about history and or a variety of disciplines, and can then connect all those dots into a pattern that works (intuitive thinking and cultural literacy). Or how one writer might seem cold or clinical to another who expresses a more emotional lived experience. How do you judge that, who gets to judge that? How do you judge where one person gets inspiration and turns that into a creative thing, over where someone else gets theirs? We are allowing the existence of “AI” to poison the well of perceived human effort. Not to mention every program or platform has some form of “AI” (which it is not artificial intelligence; that does not exist): the entire Adobe Suite, all translating programs, and most, if not all, editing programs use it. So then what are we talking about: some form of writing orthodoxy (only handwritten manuscripts, and a lawyer, or someone from of a legal witness, monitor, from when you have that first spark of an idea to the finished text?). Side note: you cannot copyright the entirety of any given text; language in and of itself is not owned, so you could only have some form of ownership of specific sentences that are inherent to the context of that text (a sentence with a specific character’s name or the name of a place specific to that storyline), but you can not copyright the generic sentence: do you want to go grab a cup of coffee.
Man, if you could copyright "do you want to go grab a cup of coffee" you could cripple romance in the United States.
LOL
That is bs on it's own, when Mfg can manipulate a finished product by more than 30% I believe and call it an original. Same with imported products into a FTZ. To have such a stringent REDICULOUS rule (...even 2% is too much) when AI assist has been shoved down everyone's throats as the 'go to' then turn around and severely penalize usage? Yeah a monkey came up with that logic. Suggest to push back the bi-polar ruling.
On the other hand, I like the idea that AI assisted works, are not original. It seperates the posers from actual disciplined daily/regular writing. Edith Wharton, guilded age author.
Care to define AI assisted?
AI assistant? Means Any writing or design of original work; enhanced, edited, co-author, relied upon for content = eventually Not an Original work. My strong but humble suggestion is don't give anyone a reason to claim your work. Refrain from using. Fk them.
How is using AI to edit work, any different to using a human editor or a thesaurus or spell-checker? As long as you use AI to mark the suggested edits, and you accept, reject or modify them, then I don't see that as an issue.
Excactly! That’s actually the loop hole. As mentioned if AI is used to ‘create’ the text, subject, storyline? That’s when this $2M deal gone wrong applies: Actual copyright - however, the laws have NOT been updated to show ‘who’ would be the copyright owner of AI generated text? That’s the dark hole real question of who owns the work? And who get’s for it.
Muddy waters. Here is a link to what the US Copyright Office has to say. It is far from definitive
https://www.congress.gov/crs-product/LSB10922
"From what I understand, via research, is that you cannot copyright AI content."
Where did you get this?
Probably fake academic work takes a little effort, too. Or fake Rembrandts. Or laboring with love over a fake wine label to create the aura of age. It’s about integrity.
He didn't copy anything. It was not a fake Norman Mailer or fake any other name writer. He used current tools, like the people before him used word processor and before that type writers.
But it is. The entirety of AI's knowledge is made up of stolen art. It's the work of all the Mailers and all the Rembrandts combined. Same with AI music. It's stolen from everywhere and everyone.
According to your interpretation almost all PhD dissertations are stolen from everywhere and everyone because they are based on preceding knowledge and studies of books in the libraries by other authors. If ingesting existing work to learn structural patterns, stylistic techniques, and foundational principles constitutes "theft," then every scholar, painter, architect, and writer in human history is a thief.
Humans are not blank slates either.
They are inspired by everything they come in contact with. Much of the tim we are impressed upon subconsciously and what we believe to be original isn't completely original. How do you xplain multiple people having the same independent idea? Liz Gilbert talks about her own theory on how she believes ideas are formed and what happens when we don't act on them.
Just like graphic design and digital music, AI is absolutely pressing on the starving artist complex and exposing the lack and scarcity wounds in creatives. Creativity is infinite.
AI is a tool that still have many misconceptions because more people have opinions about it, than experience using it.
Ok, let’s take music as an example; in the last 126 years, as a very conservative estimate, about 100 million songs have been recorded, with the same variation of 4 chords. Yet, by any measure of legal, artistic, and or cultural standards, almost none (only an incredibly small fraction, a few hundred maybe, out of millions) of them are direct copies. So then what is, or should be, the boundary between a shared cultural structure and stealing a work? And who judges that?
That's not at all the same though, is it? All those people who used word processors and typewriters still chose every word themselves, agonised over structure and meaning
But, did they? Did they run their ideas by a spouse or a friend who may have made suggestions or recommendations or came up with plot ideas that they ran with? My dad constitute an exception to having their own unique ideas? Would that disqualify someone from publishing a book? I think it's just all been blown out of proportion.
People all work differently. But most fiction writers or poets--non-fiction might be different with research assistants--do write almost everything themselves even if we take in feedback from friends or editors/agents. Those people often provide open ended feedback, not rewriting the sentences, and the final decisions are the authors. If you looked at the author's final draft before editor and agent feedback and the final book they would typically look very similar. Many of my short stories have been published with no one else looking at them except the magazine who might offer nothing more than a few typo corrections for edits.
That's quite different from having an LLM do major rewriting. That is far closer to having a ghostwriter or writing in collaboration with another author.
I don't disagree with this at all. But it is interesting that we sometimes get significant input from other people, with whom we may discuss our ideas or show them a draft of the manuscript and they may come up with great suggestions, and I don't think we always credit them as contributing authors or anything like that.
But does it mean that I should gather the opinion pof 100 friends over 5 years and 200 coffees?
Why not use LLM?
Seems extraordinary to me, the lengths to which you and others, like “Gene”, below, are going to legitimize this scheme by which a person claims to have written something that he did not write, to have created something he did not create.
Not buying it.
But everyone wished to have bought a work of Andy Warhol.... Warhol literally declared "I want to be a machine" and built The Factory around silkscreening, mechanical repetition, and assistants executing his concepts. His entire career was a deliberate challenge to the romantic obsession with manual, unassisted "authenticity." The Cretaceous period is long gone and we have been privileged that the tedious work is getting less and less. Assuming that somebody using AI as a tool did not write a book,and did not create what he wrote shows a fundamental misconception how LLM actually work.
But this is a kind of myth, no? The writer, the artist, in the studio, isolated from the world, an island unto themselves, the lone creative, and all credit is somehow theirs. Let’s take an architect, for example; they do all the initial research (where does all that research come from?) and drawings (is that style and understanding of drawing solely unique to them and only them), then there is the construction of whatever they have designed: do they do any of the labour - the carpentry, masonry, plumbing, electrical, roofing - do they do any of the manufactioning and or the production of any of those materials - did they invent those concepts and or products: plastic, ruber, the boiler, joinery systems, how and where things are mined, but in the end their’s is the name on the building. There is always a mass of invisible labour, and a supply chain behind any idea that gets produced. Or, Andy George’s six month experiment to produce a chicken sandwich from scratch.
I respect that
So, ANYBODY using AI will generate the same level of output?
No. Try to throw a chapter of say, Camille Paglia's Sexual Persona into chat GTP and it will come out as non offensive, nobody-get's-hurt kind of compliant drivel without bite and the polemic removed. Most probably chatGTP will refuse to work on it because the content "may insult" or, oh my God, "hurt other users". For example, I threw Paglia's infamous NYT op-ed of 1991 "Ninnies, Pedants, Tyrants and Other Academics" into chatGTP (without the authors name) and prompt it to "correct" it. It refused: "This content can’t be shown because it may conflict with our guidelines for creative content."
Unless you write a book which conforms to corporate compliant language rules and guidelines (like the drivel on Linkedin) you won't be able to have AI write anything with an authentic voice. It will smoothen the edges, hedge any plain statement (one view to many views) and throw in all the ideological veneer of the last 15 years. Values which America assumes as usual are universal but which come across as colonial if you live in a civilization with different value systems. A kind of white man's (and ironically now black man's) burden in a new modern coat but with the same colonial whiff and a version of Newspeak of Orwell's 1984.
American identity politics, modern diversity rhetoric, and corporate "alignment" rules have become the new moralizing export in all US produced AI platforms. When corporate institutions enforce these sanitized standards globally, they present themselves as progressive, yet they execute the exact same patronizing policing—deciding what language is "safe" or "acceptable" for everyone else.
If any random person uses an LLM, they get generic, standard output, sanitized, risk-averse corporate noise, because they lack vision, technique, and editing discernment. The quality of what comes out depends entirely on the sophistication, direction, and critical eye of the person using it.
I doubt fake works will slip by editorial teams. Any serious academic is known by colleagues in the discipline, and academics usually works with multiple authors.
No more fake than artists tracing or projecting photos to draw. It used to be a big no-no. Now, no one cares except the most obsessive old-school art teachers. Renaissance and later painters supposedly used optical devices to help with perspective; Vermeer was frequently accused of doing it. You can argue it was the technological equivalent of using LLMs for writing today. Do we care now?
Literally just said this to my husband. AI is just an $$ accessible ghost writer.
Agreed. Well said.
This is exactly where I struggle with the AI debate.
I recently ran one of my own articles through an AI detector out of curiosity. The experience was mine. The thoughts were mine. The photograph was mine. I wrote the original piece and used AI to help edit it.
The result? **0% human.**
Apparently, my thoughts + my experiences + my writing + AI editing = no longer human.
Meanwhile, publishing has used editors, ghostwriters, researchers, and entire teams behind authors for decades.
I understand the concern about someone having AI generate a book and passing it off as entirely their own. But if we're going to start judging authors by unreliable detection software instead of the actual creative process, we're heading into some very muddy territory.
It’s both comical and ridiculous: AI is now supposed to decide what is AI - a snake eating its own tail. And if AI can supposedly detect AI-generated text, it’s not much of a leap to have AI generate text that the same detectors won’t recognize as AI-generated. This makes the AI detectors a naive and temporary idea.
BTW, these are LLMs. There’s a perfectly respectable view that they aren’t really “AI” yet, but extremely sophisticated statistical language-generating machines
Drive to the Grocery Store
If you want to understand the publishing industry's AI panic, try this exercise:
Drive to the grocery store. Just a normal trip. And try to list every single place your life touches AI before you get back home.
Go ahead. I'll wait.
Let's Count
Before you leave the house:
- Your phone's weather app (AI prediction)
- Your car's route optimization (AI)
- Your smart home adjusting the thermostat (AI)
- Your email filtering spam (AI)
- Your credit card company's fraud detection (AI, already decided whether your card works today)
On the drive:
- GPS navigation (AI)
- Traffic predictions (AI)
- Stop light timing (increasingly AI)
- Your car's safety systems (AI)
- Billboard targeting (AI—different ads for different drivers based on data)
- Gas station pricing (AI algorithms)
- Street cameras tracking your location (AI facial recognition, license plate readers)
- Your bank monitoring every transaction in real-time (AI fraud detection and behavioral analysis)
- Cell tower tracking your movement (AI location services)
At the grocery store:
- Shelf prices adjusted by AI algorithms
- Store layout optimized by AI
- Your receipt's coupon suggestions (AI)
- Self-checkout scanning (AI computer vision)
- Credit card processing (AI fraud detection again)
- Inventory management (AI)
- Recommendation emails later (AI)
- Security cameras tracking your every move (AI recognition, behavioral analysis)
- Your loyalty card tracking exactly what you buy, when, how often (AI profiling)
- Store owner's AI predictive analytics about your shopping patterns
- Facial recognition at entry (some retailers now)
- Behavioral tracking (how long you stop at each shelf, what you pick up and put back)
On the way back:
- All the same stuff again
- Music streaming algorithms (AI)
- Targeted ads following you around (AI)
You probably touched AI 50+ times. Conservatively. In one errand.
Half of it is *surveillance*. Your bank analyzing your behavior. Street cameras tracking your face. Retail stores mapping your movement patterns. Your loyalty card building a profile of your entire life. AI systems making decisions about you based on all that data.
Nobody's panicking. Nobody's calling for authentication. Nobody's saying "prove you actually wanted milk and not AI milk." The grocery store isn't holding an emergency board meeting about whether your purchase was authentic.
The bank isn't issuing statements about the "integrity" of your transaction. The surveillance system isn't requiring you to prove you consented to being tracked. Street cameras aren't requiring verification that you're "really" walking to the store and not some AI simulation of walking.
The Actual Situation
You live in an AI-saturated world. Have for years. It's so embedded you don't notice it.
But the moment you use Claude to tighten the prose in your cookbook, suddenly it's a moral crisis. Suddenly we need industry standards and authentication and Pangram scans and public accusations.
The publishing industry didn't get angry when their distribution networks went AI. Didn't protest when Amazon's recommendations went AI. Didn't lose sleep when their own PR platforms switched to AI. Didn't demand authentication when grocery stores started using AI pricing algorithms (which, by the way, are arguably more ethically fraught than you editing your manuscript).
They only get angry when *authors* get access to tools that used to cost $15,000.
The Actual Moral Hierarchy
Here's what's real:
Surveillance AI (bank analyzing your behavior, street cameras tracking you, retail mapping your movements): Invasive. Ethically questionable. You're a subject being profiled. Your data is being sold. Your patterns are being predicted and monetized. This is actually worth panicking about.
Your cookbook being edited by Claude: Harmless. You're using a tool. Your work is still yours. Nobody's profiling you. Nobody's monetizing your surveillance data. This affects nothing except your editing costs.
The publishing industry is fine with the first (doesn't even mention it). Panicking about the second (calls emergency meetings, pulls book deals).
That tells you everything about what they actually value.
What This Really Means
The panic isn't about AI existing. AI has been doing the heavy lifting in publishing infrastructure for years. They've been fine with it. Profitable with it. Meanwhile, they're completely comfortable with surveillance AI that's infinitely more invasive because it doesn't threaten their control.
The panic is about *visibility*. About the fact that you can now see the tool being used because you're the one using it. About control fragmenting. About economics shifting.
So they manufacture crisis language:
- "Authenticity
- "Integrity"
- "Readers deserve to know
- "Industry standards"
- "We need to protect literature"
Translation: "We need to stay in control of what counts as legitimate."
The Test
Ask them this: Remove all AI from your publishing operations. All of it. Your PR platforms, your contact systems, your recommendation engines, your distribution networks. Shut it down. Do it today.
They won't.
Because they don't actually believe AI invalidates work. They believe it invalidates their gatekeeping power.
The Final Hypocrisy
They're using Pangram to accuse you of using AI.
Pangram is AI.
They're using an artificial intelligence system to detect artificial intelligence use, calling it "proof" that your work isn't authentic, while they themselves remain completely comfortable using AI at every level of their own operation.
They won't give up the surveillance AI. Won't give up the profiling AI. Won't give up the infrastructure AI. Won't give up Pangram, the AI detection tool itself.
But you should give up Claude. For integrity. For authenticity. For literature.
Drive to the grocery store. Notice everything running on AI. Notice nobody panicking. Notice it's working fine. Notice the only people asking you to sacrifice anything are the ones using more AI than you've ever touched.
That's not a crisis about literature.
That's a tantrum about losing gatekeeping leverage, dressed up in moral language and powered by the very tools they're accusing you of being corrupt for using.
Agreed! It's the death rattle from an industry that can no longer buy the attention individuals can generate on their own. They started offering book deals to anyone with an audience because they cannot compete in the attention economy .They literally have to purchase their audience, NDA the hell out of the people who bring the audience, and buy their IP/Name/likeness etc to profit in perpetuity. It's gross and the individuals inside that industry need to re-purpose their skill sets into something generative instead of extractive.
I think you're correct that LLMs let people generate their own content and self-generated content dominates the current "attention economy." This is a big threat to AI-assisted writing too. Why pay for content someone else didn't write when you can generate your own tailored to your specifications?
Every social media influencer above a certain amount of followers is being offered a book deal. Steven Bartlett has pivoted his media company to signing deals with Podcasters that already have established audiences and streams. Companies are not buying content, they're buying eyes in they same way they're buying data.
Anyone seeking external validation and who believes they need someone else to prop them up to be 'successful will sign these inherently bad contracts.
Everyone thinks the AI conversation is about creative agency when it's really about self worth. Creatives have notorious chips on their shoulders about their creative worth, which is why the 'starving artist' mentality persists and why many will continue to suffer for their art and feel victimized by this technology.
The question here that sorts all of this is: do you have a non-ai, human centred process you follow to produce your work? If you do, then in the production of your work, as long as you only use AI to augment your process, not replace it, then that means the work is still yours.
I so agree with this and for people who are really doing their own work and then using it for editing. The programs that read writings for AI content have some issues as well.
Hi Nicole... This conversation definitely extends beyond AI. The Tech position/push is for ...fair use... as purportedly, despite the strong use of AI to create a piece of art, AI produced work is being defined as belonging to the public domain. I am not certain if I agree with this legally, as the very definition of copyright applies not to ideas, but to the expression of those ideas. Trillions of bits of data that belongs to us all is the fuel that AI programs use to produce creative works.. The content belongs to everyone, and this is an issue that demands our attention. Without our input, there is nothing for AI programs and systems to use. This raises several issues, and we should explore how contributors to AI can be enriched by making possible the creation of these works developed/interpreted by AI programs and systems.... Who knows... Maybe we can create an economically just society.
Falade, lost me at “Three Black authors received major publishing deals this year, and all three later had deals cancelled or disrupted following suspicions about AI.”
If there ever was a time not to be played the race card this is it.
Tell me you don’t acknowledge systemic racism without telling me…
Wake it up! These people say it so casually they don’t even realize it lives in their very souls.
You can acknowledge systemic racism while at the same time calling bullshit on using the specter of systemic racism to pull off a con job.
You’re telling me that the alleged-ai-assisted authors “used the spectre of systemic racism”???
My point was that there’s a very good chance that non-poc authors have been using ai and getting away with it, and that they weren’t caught because of their privilege and the default level of trust granted to them and not granted to people of color.
Yes, Falade did exactly that. Did you read his response?
Listen, in many aspects of American Society POC certainly face discrimination. I don't deny that, but in both publishing and academia, they are often given much greater deference because everyone wants a diverse roster and nobody wants to be accused of racism. You need to realize that some environments/disciplines become ideologically captured and the larger power dynamics get interrupted and sometimes reversed.
I think both black and white authors are probably using AI unethically and using it at the same rate, but there is not anywhere near enough evidence to prove that racism is the cause of this publisher pulling the book.
According to your flawless logic, catching someone red-handed isn't actually about what they did, it’s just a tragic case of unequal surveillance.
Apparently, when non-white authors get caught using AI, it’s not an ethics violation; it’s a systemic injustice because we haven't caught enough white people yet! (Am I understanding your logic"?)
Your defense: "Sure, my client cheated, but what about the hypothetical white people who might be getting away with it?"
You're lauighable logic is a masterclass in shifting goalposts! As for your tearful claim that white authors are floating through the publishing world getting "showered with praise" for dropping AI prose into their manuscripts, here are 3 reality checks for you:
Mia Ballard: A white horror author who's novel "Shy Girl" was canceled by her publisher Hachette following an investigation into her use of generative AI. Her publishing career was essentially nuked overnight. Where was her magical "privilege" shield?
Stephanie Johnson: Disqualified from the $65,000 Ockham Book Awards because AI was flagged on her book's cover, a cover designed by her publisher's design studio that she had no part in creating. Total zero-tolerance, immediate disqualification.
Elizabeth Smither: Was stripped of her eligibility and booted from the exact same $65,000 Ockham Book Awards for the exact same reason, an AI-generated cover image she didn't even design.
Turns out the publishing industry is nuking anyone caught using AI, regardless of what skin colour is typing. But please, keep weaponizing "systemic racism" to defend literary fraud, it’s much easier than admitting people should just write their own books.
Universities have actually stopped having students run papers thru AI checkers because they flagged papers that students could prove were not AI generated. If I’m writing now I’m saving every 5 minutes and emailing copies to myself so I have receipts
It's chaos in academia. No clear standards. Varies by university. A real mess for everyone.
Unfortunately not enough. Saw on reddit how many lost their scholarship or was downright kicked out or nearly kicked out because of that dumb shit. How are you gonna up and accuse someone of using Ai while USING AI TO SEE IF AI WAS USED OR NOT?! They even said at this point you have to literally freaking record yourself writing and typing out everything and still probably won't be enough.
My god please dont be someone who has a strongly preferred way of typing or have aspies, otherwise, your done.
Even with the written manuscript, notes section with your jumbled up notes and a new note with the story chronological in order and a final typed out version that's done after your done writing the manuscript IT WON'T BE ENOUGH. All because of some bitter lazy cunt.
I graduated from grad school in 2017. I used Grammerly as a thesaurus but it would pop up and ask if I needed help. I never took it as I’m a strong writer but could someone be expelled for saying yes?
We were supposed to submit papers to be checked for plagiarism but most professors refused. The experienced ones always said they learned students authorial voices very quickly, I remember one prof saying within the first week that because of my large vocabulary (I’m still not sure how she knew that based on a few classes) I was most likely a good writer. AI is making us doubt our own brains
On my research I saw a number of them got in trouble due to using it or others similar to it – But it was when they plugged that Ai bs into it or was waaaaaay too formal and proper. I wouldn't use any of those assistance texting because it just adds to them claiming you used it. Type it out leave and have a billion errors and go back and then clean it up manually – Yes, it'll take a billion years, YES it will drain you out as i do when I go and have to redo my short stories and novels and screenplays, but its so worth it in the end and helps the flow in your brain btw.
Plus also noticed, the more I use spellcheck the more I forget how to spell and use words and punctuation which weirded me out BIG TIME.
And I saw a video on yt this girl called readers out on throwing tantrums when writers use big or unfamiliar words; They wouldn't last a few minutes with a Cormac novel or reading PUSH. You can't use big words, can use these ( – or – or —) without some idiot screaming AI! AI! AIIII!!! Cant be expressive and highly descriptive then getting right to the point like NCFOM – Like we can't do anything without some idiot saying its Ai. Remember the good old days when your sole worry was someone saying you stole your idea or work from someone else or ripped off so and so.
And sounds like she was either sizing you up or maybe you were one of a very few to not only do the work but love it and not complain or half ass it? Its so stressful it really is. Gotta basically censor ourselves out of fear of one idiot accusation without proof. They already don't read our manuscripts or assignments or anything as is; Be lucky is they skim through it. Don't know how they can wake up and call themselves teachers or publishers or editors without even looking at the damn work we produce but come to a quick; And usually incorrect. Conclusion.
Its either dumb yourself down and don't stand out or... be accused and ruined for life... Sorry for the rant btw broskie.
I hear this argument a lot: "How are you gonna up and accuse someone of using Ai while USING AI TO SEE IF AI WAS USED OR NOT?! " and it makes zero sense to me. Students are demonstrating competency when they write a paper. Professors are not demonstrating competency when they evaluate the papers! They are trying to determine if the student has met the objectives of the assignment and part of that is making sure they don't cheat.
You missed the entire premise: when a student uses AI to write their paper, they aren't demonstrating competency at all.
An assignment is supposed to be evidence of the student's internal understanding and critical thinking. When an LLM generates the arguments, the paper proves nothing about what the student actually knows.
Claiming a student who used AI "demonstrated competency" is like claiming someone ran a marathon because they took an Uber to the finish line.
You have completely misunderstood my post. I was replying to onnoemelijk above.
Eh, it fits the same mold that racial prejudice can be sneaky. You don’t have to be overtly racist to scrutinize people differently without realizing it. When you see a pattern of black authors getting x 80% of time and others getting only getting y, that could be why. However I agree three is simply not enough datapoints to jump to that conclusion and that makes it feel like card being played to rally people to their defense, not a real accusation with evidence and merit.
(a) Falade doesn't know how many Caucasian, Asian, Indigenous, green, and purple people have been rejected or had their advances rescinded because of AI use; hence, to your point...
(b) Without a full picture of the publishing industry's purity testing of manuscripts, Falade jumped to racism, likely in an attempt to shame his publisher and garner public sympathy.
Falade's accusation of racism is a blatant deflection. By substituting outrage for evidence, he undermines his own credibility. He needs to do the hard work of proving he wrote his novel without AI, rather than hide behind unfounded claims. Crying "wolf" only trivializes real discrimination.
You may be correct, but the term 'the race card' is annoying and generally used by racists.
Good thing you only speak for yourself, but I gotta hear this… please explain why you believe “race card” is used by racists and provide the data you used to come to your belief.
If you spent any time on the cesspool X or see any racist Republican talking points, you'll hear it at some point. I think you know this, and you're just trying to be coy. If you don't know any of this... where the hell have you been? I suggest you do your own homework for the data.
RE: “Do your own research.”
Ah, the classic battle cry of someone whose entire argument collapses the moment you ask for a single, solitary source.
Congratulations, by the way; you’re officially the first person I’ve ever seen attempt the intellectual gymnastics of claiming the "race card" is only played by racists. Masterful projection. Truly, 10/10!
Try again when you've found a better script.
*I commented this further down on a similar reaction but felt it was a valid take to repeat once more if nothing but for a different perspective*.. You don’t have to “race-bait” to question any repetitive accounts of a similar nature. And news flash,people, like yourself are not easily lead to believe that someone is being discriminated against,especially for being black in anything! This current climate of perception vs reality in regard to racial biases provides not a single black artist, author etc any favors. In fact,as evidenced even within your own response, it elicits more scrutiny and negative reactions than anything with empathetic understanding. I do not believe this author was trying to use an imaginary “race card” here, but was concerned with his own findings and like most black people in general, truly hope and even fantastically wish that to not ever be the truth. Remember, just as this author openly questioned the process based upon his own meetings with everyone involved,you chose to question the author’s validity based on your own limited understanding of what he was told behind closed door meetings where transparency rarely exists beyond what is recorded.
You conveniently ignored “Falade doesn't know how many Caucasian, Asian, Indigenous, green, and purple people have been rejected or had their advances rescinded because of AI use.”
The publisher was willing to pay $2 million to a black author, but now that Falade’s use of AI has been questioned BY HIS REPRESENTATIVES, NOT THE PUBLISHER (Funny how everyone missed this fact - Read the article; it’s clearly stated in the 2nd paragraph.), it’s a “race issue” rather than an issue of “AI usage”?
Explain what a person’s skin pigmentation has to do with their being caught doing something they shouldn’t have done.
Is the issue at hand a race issue or an AI usage issue?
I find it most telling when someone defaults to the “ I don’t see color” or “green or purple people” explanation.Let’s be clear here, there are no imaginary green or purple people. We have what we have and that’s diverse enough without making things up to express a lack of intolerance towards a specific race/culture. I have no idea what the author knows in regard to how many other authors, and their ethnicity have been issued the same treatment. And I’d gather that neither do you.You are simply adding your opinion based on your experience(s).And I shared my perspective as well. But let’s wait to see if all the details come out first before we write someone off or show callousness towards how they present their story. The same goes for his reps.I can’t say they were right or wrong in the situation yet,so I won’t judge either way!
Appreciate the zoological update on imaginary colours, followed by a full essay dedicated entirely to saying "I don't know."
Here's what you do know: Falade's agents walked away from a guaranteed $300,000+ payday.
Based on my experience, people don't walk away from $300K without a very good reason; hence, Falade's agents walked away knowing, with absolute proof, that their client had used AI to write his novel.
Nobody cares what colour Jerry Falade is; it's the character underneath that makes him so disappointing, as you are also trying to excuse someone's lack of ethics because of their colour.
One would reckon that if, as the trifecta of Falade, HM Wolfe, and Mia Ballard have proven, if it were not at least partially racially motivated, there would be more public news stories, if not more private industry chatter that was worthy of industry-press reportage, about white writers losing opportunities or having already published work cut for the same, especially white authors who have admitted to its use several times. To wit: if it were in fact at least aiming for parity, James Frey would have lost out on the opportunity to publish *Next To Heaven* at least in 2025, if not in 2023 when he said verbatim in an interview that he doesn't care how readers view AI use in his fiction.
The public narrative isn't one Falade constructed. The mere fact that publicly only Black writers have not only been punished but had their punishment publicised would still be an issue even if one can prove parity, because then the question would be why the industry, if not the press, would not be concerned with portraying the issue equally.
Brandon, conveniently glossed over a massive detail: the publisher didn't cancel the deal, Jerry Falade's own agents voluntarily pulled the manuscript and fired him as a client.
Marc Gerald and Sandy Hodgman willingly walked away from a guaranteed $300,000+ commission on a $2M+ multi-publisher bidding war because, in their own words, they could no longer "authenticate how the manuscript fully evolved" after Falade’s story about his process kept changing. Agents don't burn a seven-figure payday on a whim unless there are red flags are glowing in the dark.
This isn’t a unilateral industry ban; it is fundamentally an AI usage and disclosure issue. Every publishing house, and every literary agency, has a completely different threshold for what level of LLM involvement they tolerate, ranging from "zero-tolerance" to "fine for research, off-limits for drafting."
When your own literary representatives pull your book from an active $2 million auction because they can no longer vouch for its authenticity, the problem isn't the publisher's policy, or some "ism," it's the origin of the manuscript.
That detail is not relevant to the argument I am making. It's kind of rich to say I am conveniently ignoring something when you didn't respond to the core of my message.
To put more bluntly: if Falade *did* in fact fake the text by AI or other means, he does deserve to be punished; and if his agents were gun-shy for any other reason, then the specific redress issue Falade has is specifically with them. The individual agents representing anyone are allowed to say no for any reason, that's how the industry works; and Falade is welcome to defend himself as anyone else would be in this instance, so the only part that matters is the evidence himself — evidence Falade insists he has and they never asked for, which is why I imagine he's planned to take it to court.
But it is undeniably still noticeable that consequences like Falade's, regardless of the arbiter, are based on assumptions of dishonesty that have only been publicly documented and discussed when it affects writers of colour under suspicion, when white writers who confess to the same deed we agree deserves consequences face none.
What part of “his own agents pulled the book from a $2M auction” isn't sinking in?
You really think veteran agents walked away from a guaranteed $300,000+ payday without absolute proof Falade used AI? That's not how commission-based businesses work.
And spare me the whataboutism. Playing "Well, what about other writers?", which is childish, completely ignores the fact that every agency and publisher has its own distinct AI policy, from strict zero-tolerance to research-only. Falade was held to the rules of his agency, not a fictional industry-wide standard.
You dont know whether anything you wrote is true but wrote it anyway as if its gospel.
What part do you believe isn’t true?
Bingo. Today, what professional in publishing or anything else wants to be accused of racism? Are you kidding? Those folk did backflips and other contortions before they brought themselves to withdraw a black author’s submission AND forego their fee.
Right on. It's incredibly tiresome.
We don't care where he lost you Nick.
Correction @chandra …
“You” don’t care.
Since you don’t speak for everyone, just yourself, using the word “we” is a lie.
If you’re going to be easily offended, bitter and angry then at least don’t lie and understand you only speak for yourself.
P.S.
The topic at hand is about AI usage. Trying to make this about race is nothing more than gaslighting to deflect from the Falade having used AI to write his novel.
No, I said what I said, we don’t care. And no amount of name calling is going to change that. Without knowing any facts of the wrongful accusation against this author, you’ve already written it as a fact that he did use AI. As I said we don’t care.
You can stomping your feet in anger all you want, you won’t change the fact you only speak for yourself.
Can you read?
THE AUTHOR’s REPRESENTATIVES, NOT THE PUBLISHER, pulled the manuscript, therefore in all likelihood Falade used AI. (This is clearly stated in the 2nd paragraph.)
In any case you are obviously bitter and angry and rather stay that way by looking for unsubstantiated reasons to stay that way as opposed to staying on the topic of AI usage, which has nothing to do with skin pigmentation.
I agree. I felt the same. This is about AI use in claiming ownership of one's writing. Why did the race related assessment get in here, and if that's a valid observation, then that's just very sad state.
100%
This isn't a race issue; it's an AI issue. His reps pulled the manuscript over how it was written, end of story.
No surprise that he lost you…
He doth protest too much…?
Same. I immediately thought it seemed like a desperate move by a guilty man.
https://charlesdunnchuck.substack.com/p/around-the-kitchen-table?r=13uiot&utm_medium=ios
Some research does show that AI detection software is pinging non-native English speakers, writers of color, and neurodivergent writers at higher rates than native neurotypical white Americans.
It's not just a black issue - it's a problem for many minorities.
I guess you can find excuses using “some reserach” to try and justify unethical behaviour.
If you read the article you then know that it was his agent who withdrew Falade’s manuscript, not the puiblisher. Therefore, if his agent didn't have proof that AI wrote part of Falade’s the book, he'd never have walked on his 15 - 30% or whatever a literary agent receives.
This isn’t a race issue, this is an intergity issue.
Regardless of who pulled it or why, (and yes, I read the whole article?), if it was because an AI detection software pinged some of it as AI, there is plenty of research to support the prejudice of AI detecting software. Of course "some research" is more correct than "all research". Very few things are absolutely conclusive. An agent calling a false truth does not make the truth any less false because it was not done by a publisher. They both have a lot of money to gain and reputation to lose. The unknown: is the agent inadvertently lying in an attempt to avoid controversy that may lead to even larger losses, or is the author lying? We would need to know what the evidence was, how it was gathered, and determine how often that method fails in independent testing surrounding prejudice. I tend to trust a human over AI, solely based on the vast failings of AI. AI is great at managing data, but not so good when it comes to "forming conclusions" or "creating". Using AI to detect AI seems incredibly fallible on its own, due to a lack of objectivity, independence, and neutrality. Have you been following the research and testing of some of these AI programs? The data says it all - they are incredibly fallible, and on some cases outright dangerous.
Groundbreaking analysis, Tina. So because AI detection software isn't 100% flawless, an agency isn't allowed to protect its own reputation?
I guess that if some research says AI detectors are bad, then by all means, let’s completely disregard Falade’s now-former agency’s right to make decisions for their own business. Because clearly, an agent making a business call is basically the same as a grand publisher conspiracy.
How silly of everyone else not to realize that dropping a client over flagged content is a deep, sinister motive, and not just, you know, standard operating procedure for every manuscript submitted today, regardless of who wrote it. Silly me, I thought reputation and business standards mattered!
P.S.
At least you wrote “Very few things are absolutely conclusive,” yet here you are trying to make what happened to Falade conclusively about racism.
I'm saying that the author could be telling the truth or the agent could be telling the truth. Who are we to decide between the two without further testing and a knowledge of AI detecting software's constraints.
The eventual reality will be as self publishing.
I believe it's a discrepancy that needs investigation. It's relevant; it's part of the debate. It cannot be side-stepped. Nor should it be.
What “discrepancy” are you referring to?
Did you read the article?
Here’s how the 2nd paragraph begins: “In an email to publishers, one of his representatives wrote….”
It was Falade’s reps who questioned his AI usage, not the publisher.
Waa
If anything this is a VERY valid reason to play the “race card” it has only been black authors caught up in this AI witch hunt.
"Other people haven't been caught yet" isn't a free pass to write a novel with AI. Your logic, or should I say, a lame attempt to frame Falade getting caught using AI as a racial issue, is misguided. A literary agency protecting its business standards from AI content isn't conducting a "witch hunt"; it's basic accountability. Being caught violating industry standards doesn't inevitably become a racial conspiracy just because you want it to be, just because others, with the racial profiling you'd like to see get caught, haven't been caught yet.
Speeding is still illegal, even if the cop didn't pull over every single person on the highway.
Great discussion.
I'm stuck on this question: if a book or any text is appealing to a reader, does it matter whether ti was written with the assistance of AI? That's a different question than being dishonest about whether it was written with the assistance of AI.
I think it's inevitable that sometime soon a book will be written with AI, that fact will be disclosed upfront, and the book will be both popular and well reviewed.
To me the dishonesty is the problem, not the use of AI.
This comment was NOT written with the help of AI!
You’re right. It’s the dishonesty in passing the work off as one’s own.
I'm equally as offended when I find a book wasn't written by the advertised author, but a ghost writer.
Why? If you enjoy the book what do you care who wrote it? I'm just asking this out of a curiosity because I don't buy books because of authors I buy them because of content.
I'm referring to the statement about dishonesty. We can't make deception statements about AI without being brutally honest about how publishing actually works. Conversations like this shake out this information. A published authors work is rarely untouched by many other hands, voices, POVs etc. So using deception as the argument about AI written pieces is thin.
In the end it doesn't matter as long as it does what it's supposed to do just like talking about it makes it matter
What of this
Using ai does not make it dishonest its still there work gen ai would never get that far
not surprised you don't know he difference between "there" and "their". AI work is stolen work. Point blank fking period.
Ai work is not by any deffinition of theft classified as such
Full stop
Come back when you receive proper education on the matter lies have no place here
Yes I think the dishonesty is key. And it's worth noting that big publishing houses published several openly AI co-authored books. A few got major publicity that any author would dream of. But... the sales weren't there. People don't want to buy AI work, at least if they know it is AI. That might change in the future.
I'm curious. I did a research project recently among landscapers about how they use social media to help them do their jobs. They all use YouTube videos while on the job to help them do the job right the first time. I said, "If someone asks you whether YouTube came up with the idea or whether the idea is yours, what would you say?"
They all said they would say, "I figured it out." I kind of get that because they took on the knowledge and actually did the job. Maybe this isn't exactly the same thing but it's an interesting moral analogy.
That is interesting. Somehow I think It’s a different dishonesty but I can’t explain why.
It is different but -- to me -- the shared aspect is taking credit for something that really came directly from an outside source. I think there's an aspect of human nature in play here.
If I read you right, you're saying when on a landscaping job they would look at how-to youtubes to execute a task? That doesn't seem any different from taking a guitar lesson or reading a poetry craft essay. No one expects a landscaper to come up with a new method of planting a bush or the idea of using gravel.
If you mean they copied someone else's designs entirely then I think the difference might be the type of work. A prestigious landscaper hired to design a prominent botanical gardens would probably be judged for ripping off someone else's work in a way few would judge your average landscape job. Similarly, as a writer I have a much higher--and just different--standard for myself when publishing a novel under my name than I do writing anonymous freelance work for cash. I imagine most fields are that way.
So, a human writing coach (teacher) is OK, but using an AI writing coach (teacher) is not OK. Is that correct?
A human writing coach will not give you the text and say “now go publish this as your own.” They will more likely prompt *you* to help you generate interesting work, or describe what they are seeing in your work so you can assess how well that matches up with what you wanted to put forward. I guess one could use AI that way too, perhaps, with the right prompts, and most likely this is a better analogy to the landscaper example of using YouTube to watch someone else and then replicating those steps with the material at hand. But the minute you copy and paste and don’t disclose it, this is not “coaching” and it’s time to stop fooling yourself that this is independent artistic work.
You can tell AI not to write your story. You can use AI to re-compose a poorly written sentence. You can use AI to learn more about the location of your character for contemporary fiction, how your character might live, get around or eat. I also use Google Maps to learn more about the location. I ask AI how a specialist such as an intelligence service or police might word something. Many other details too.
I also think “self publishing” is not self publishing if you pay someone to do it for you. Same for book covers. Expensive book covers are not your own work. Make it your own by making your own. It is free.
Information is out there to be used, and we cannot keep citing the original inventor/creator of such information. Otherwise everything we write here must end with: this text is made possible by Grunt Argh, the inventor of English, by Jada Ba, the inventor of writing, by Ming Xu the inventor of paper, Christopher Sholes, the inventor of the typewriter, ……….
We should be able to own copyright on our ideas however that idea is manifested. Just because I brief eg a photographer to produce a photo to my specific brief, it should bey.copyright. he wouldn't be producing that photo of it wasnt for my brief.
Likewise AI.
It doesn't just boot itself up and produce a photo or book because it had an idea. The day it does, and that original idea is demonstrably its own, then it can be said it was its own idea.
I totally agree with you. Who gives a crap as long as the reader enjoys it? We've gotten so pompous and self-righteous like we're special artists and we have a secret that nobody knows about. Poppycock.
THIS!!! To me, this is the actual question. In simple terms, can an AI story actually be good, not in terms of prose or flawless grammar but is the story captivating. If an LLM is basically a prediction machine, can it take a story somewhere new?
I don’t think so.
I think the article covered this topic a bit. There will always be a general publishing industry that is meant for “consumption” rather than “absorption” just like there are commercial movies and Oscar worthy indie movies that will take special taste to watch. One can be a consumer and appeal to any form - actual human input or LLM driven slop.
As soon as you recognize that something was written with the assistance of an AI you get stigmatized and you lose readers. Even if you worked for months, rewritten the book 8 times, and used AI for research, analysis and proofreading only (it can’t do quality literature writing).
People will continue to be “dishonest” for as long as there is “stigma”.
Let’s fix both problems, no stigma, no dishonesty 😉
obviously it’s a question that concerns copyright and who gets paid and how much.
it’s lovely to ask abstract philosophical questions about use of AI — “why does it matter?” — as if the only concern is the quality of the prose.
the quality of the prose is not the issue. “who gets paid” is the issue.
everyone loved Falade’s book. the quality of the book was never the question, the quality is not the issue in the discussion at hand.
I see some commenters asking "Why does it matter if it's AI?"
If the AI authors really believe that their work is good and important, and if they have no issue with it being AI generated, then let them declare it as such. If it didn't matter then they wouldn't pretend that they created the work themselves.
I estimate that around 4% of the stories we received in our recent short story competition were created with, or with the assistance of, AI. I believe we eliminated all such stories from the running, because we were hyper-vigilant. We had a large prize fund, so we expected the scammers to come calling. Really, it wasn't very hard to spot them. There are certain tells, like an evenness to AI writing that you don't get from an authentic human writer. I guess it will get harder to pick them out in the future, which is troubling.
And i bet money ppl you didnt pick out or consider ai did use ai
For books if you know how to use ai noone will ever know but you need to put in work at that point its your book not ais
I think the real question here is how much work do you have to put in? Many people are spamming AI generated text with essentially no human work at all. It can be automated easily. Then on the other hand you can have heavy involvement that takes as much work as any pre-LLM method of writing. The question will be where is the line drawn, no?
50%
AI can't creat things that are NOT ALREADY EXISTING. AI ISN'T ORIGINAL. IT STEALS FROM EXISTING WORKS TO "CREATE" A WORK. IT IS NOT ORIGINAL. IT'S STEALING.
Incorrect it formats words using data in a similar fasion to how we form words
Every word we say was already said 1000 times
Learn how to use ai before you join into an adult conversation about ai
Proper use of ai allows you to create original content
No, it doesn't "format" words like we do. It's not creative, it's *predictive.*
Nope!
Your reply says you dont have any knowledge on this and probobly should stay out
Wrong again. I have a considerable degree of knowledge about it, having previously worked in the tech industry for 30 years. I'm also a hard-working writer and an experienced editor. So I guess you could say I cover all the bases in this discussion.
My point remains: if work has been created by AI then let the authors declare it as such. The fact that they don't tells us everything we need to know about the degree of awareness they have that they are not creating authentic material. By trying to pass it off as human-authored work, they are committing fraud, even if it's only in a moral sense.
Let them use AI; fine. But if there's no shame in it then let them be honest about it and say:
"... and I wrote this novel using AI."
No you dont and if the author did atleast 50% of the work there is no need to disclose it
Claiming you have knowledge is debunked by proving you dont have knowledge
Tools are tools
I have no problem declaring that I used AI for my book. But I didn't just write a prompt in there and say hey, write me a book about whatever this topic is. I used it to help me develop an outline and then I modify it myself and then I use it to help me do research to find things that I would have to do myself anyway but why not use a research assistant for that. For example, I could spend weeks searching Google and read it and Facebook and other areas to try to get information about a topic that I'm working on for my book when an llm can handle that in a matter of a minute or two. I write the paragraphs. It's my idea and my thought process. And I let AI help me with making sure that I'm consistent across the board. For example, I loaded a couple other books I wrote into ai and I had it developed what's called a DNA guide which standardizes my sentence structure and my word structure and grammar and other areas that are unique to me. And then I use that to test each chapter to make sure that it follows those standards. What's the difference between doing this and sending it out to an editor and letting an editor go through and do the same thing? In summary, I don't think that it's right to allow the AI to generate the entire book from a simple prompt of write me a book about something, but I don't see any issues with using it in order to assist with research, standardizing grammar and sentence structure and the rest as long as it's still my voice.
I have no real problem with that use of it, either. For me, the issue is when 'creators' use AI to generate the paragraphs of a story/novel, or the stanzas of a poem. Research is perfectly legit with AI, it's great at that stuff, though does tend to throw in incorrect information from time to time, so verifying via other sources is probably a good idea. Dictation tools like Wispr Flow use AI; again, it's perfectly legit in my opinion to use such tools.
Just don't (and I'm not talking about you here, just in very general terms) get it to write a story/novel/whatever and then tell me that [insert name] wrote it themselves, and then call that art.
Or do, because it's a free-ish world.
To be honest, my real issue is when-AI generated material ends up in our submissions queue. Genuine artists are vying for limited publication opportunities for their work, and I feel there's something immoral about asking an LLM to create it and then claim it as one's own. It's like me buying a painting then telling my friends that I painted it.
Best of luck with your writing!
I agree with you 100%. I'm a data scientist and I do a lot of work in AI engineering and agentic development. I use AI extensively in research and statistical modeling and even report writing. But in every single case, I am the human in the loop. Nothing gets posted or released until I have had a chance to review it personally in the same goes for my book writing. I wrote a book called "service dogs for mental health" and it is done okay and I used AI for a good part of the research. For example, when I had to really get into the Weeds on what the Americans with disability act said, particularly titles two and three, it was easier to use AI to generate the language and the citations, but I had to go back in and check those myself to make sure they are accurate. In this newest book that I just wrote, there's 101 citations and I committed to the publisher that I would attest to having reviewed every single one of those personally, meaning pulling the book or the article or the publication or whatever it was and verifying the context, the content, the page number and everything else related to the citation. I just think that there's this blanket fear of what AI is going to do to the artistry if you will of authorship and I think it'll become quite apparent the difference between those that have a powerful human in the loop model compared to those that rely on autonomous writing.
I'll add, Frank, that I use AI for all sorts of things, from recipe recommendations to understanding the rules for the calculation of my golf handicap, to creating VBA macros for Excel. I don't believe it's a bad thing in and of itself, in other words.
Professional writers write to earn money. Publishers publish to make money. Agents, editors: in it for the money.
Current copyright law only protects works creates by HUMANS. there is no money to be made publishing a book that cannot be protected by copyright.
MONEY people.
everyone agreed that Falade’s book was sensational. No dispute about the quality of the book. $2 million dollars. that’s a lot of clams.
but publishers aren’t going to publish a book that has a copyright issues, where possible AI use puts the copyright under a cloud of illegitimacy.
What strikes me in all of this is how much the industry is still operating inside a causal worldview — where authorship is treated like a chain of custody, creativity is imagined as a single point of origin, and the fear is always about what a decision might cause downstream.
But for the reader, none of that matters.
The “man behind the curtain” isn’t the site of value.
The value happens entirely in the consciousness of the receiver.
Every creative act — painting, filmmaking, symphonic composition, novel‑writing — has always been collaborative, derivative, and emergent. Da Vinci had apprentices. Beethoven needed an orchestra. Spielberg works with hundreds of artists. A play isn’t the playwright; it’s the performance.
The work is never the artifact.
The work is the experience the artifact generates.
That’s why the AI panic feels misplaced. It’s trying to force a square peg (a distributed creative reality) into a round hole (a causal model of authorship). It’s the same chicken‑and‑egg problem: a paradox created only because we insist on linearity where none exists.
What’s really happening in publishing right now isn’t “AI disruption.”
It’s transition dynamics — the field tuning itself in real time as old assumptions about creativity collide with a new reality.
And the only thing that ultimately matters is the coherence of the field between creator, work, and reader. Everything else is noise.
While you can take this stance, it is obviously false to claim this is the universal opinion of readers. Many readers do care about questions of authorship, plagiarism, falsification in memoirs, etc. We wouldn't have these scandals otherwise!
My response: People often ask whether AI threatens the integrity of creative work.
For me, the distinction is simple: AI can assist with expression, but it cannot originate consciousness. “Words don’t mean, people mean.”
Only readers who think they can tell the difference. (And probably mostly failed writers)
So many contributions to the world happen from standing on the shoulders of others. But in our individualistic culture we're encouraged to take full credit for the latest creation without acknowledging the work of those before us that allowed us to progress our ideas. Maybe big stories like these will move us towards full disclosure and acknowledgement, which might not be a bad thing.
In terms of using AI for research I don't see how this is any different to using other digital tools for research (google, bing). Of course writers are using AI for research.
As others have mentioned in these threads, using AI screening to identify AI content is problematic. @ravencarriesfire wrote an interesting post on this topic: https://ravencarriesfire.substack.com/p/pangram-is-a-pentagram
We all stand on the shoulders of others, it is true and should always be acknowledged. I think this applies to knowledge of the human kind and be-ing of the existence kind as well. This AI morass is opening so many lines of enquiry and confusion it seems we will have to work it out in the commons, together, writer to writer to reader to thinker…..etc. I often think about the uniqueness of creativity and the direct outcomes of coming into contact with ‘real’ creativity. What is real creativity anyway? It feels as though we are treading water on a long line of ancestral knowing and in many ways rehashing the rediscovered and the previously known. One thing is for sure: the days I feel genuinely inspired by what I think is unique creativity is always brought to ground by the fact that somewhere someone has had the same thought at some point. Makes me feel like a pop-up consciousness @Jordan Clist.
Sorry but your final sentence sounds like noise.
That's because the comment was AI generated. (according to Pangram)
We are really living in a state of paranoia atp.
It did sound way too structured to be a comment on someone's substack, too much like an essay.
As a counter example, my kid was a very avid Harry Potter fan. After the writer started taking certain positions things changed, especially when those positions were used in attack mode. They will not do anything that enriches her.
It is not solely on the text that enjoyment and interest lies. Readers are a (diverse) community.
I can appreciate that. But how much of that is learned behavior? Or peer pressure? We enrich all kinds of people whose views don’t align with our own. But I applaud the integrity.
I'm not sure if I understand your comment. They are not enriching the author of the Harry Potter series. They will buy nothing connected to it. No streaming, replays, re-releases, not a penny directly or indirectly. [And neither will I].
This set of dictums is riddled with holes like "A play isn’t the playwright; it’s the performance." That's illogical, because whoever claims that a play is the playwright? A play is the creative product of a playwright and exists whether it's performed or not. It is an artifact. The work is ipso facto an artifact.
Of course the artifact exists. It’s like the idea that, “If a tree falls in the woods with no one around does it make a sound.” If a playwright is just writing the play for the sake of writing the play, it might as well be a diary entry. If the play isn’t to be read or performed for an audience what purpose does it serve? The communication process (of which all artistic expression is at bottom) is incomplete without the sender, the message, the medium, and the receiver.
So fking what if it had some ai
Was the book good was the book original
Thats all you need to ask
Probably not original, since it would just be a compilation of a whole bunch of others authors' work
How much do you think has been written from 1900 to 2026, including articles, books, magazines, journals, plays, films, blogs, and now podcasts? Now, out of all that, how many sentences do you think exist in all that writing? Well, it is somewhere between 50 and 100 trillion sentences (that is only the last 126 years). Meaning, depending on what you are writing about and the style you are writing in, 10 to about 30% of your sentences have been written before. There are only so many ways to describe the everyday experiences or the varying structures within maths. Computational: any given language has what seems like endless variations of itself; the problem is the human experience does not.
Which is why using ai still creates original content
But to get good content with ai you still need to put in the work thats how it becomes yours
Know how to use ai you can better your work
If so much work has to be put into it regardless, why not just... write the book?
They did write the book ai is a tool they used ai they still wrote the book
Tools are tools thats how it works in this world
When using ai it's not writing a book, it's formulating one.
Well yes, that's obvious
What is obvious, and why?
Nothing is original
The perspective or perception (the intention to preserve an idea or an artifact as a witness) turns something that is yours into something original for you, and it could make someone else think or see in another way.
The content entirely is original but you will always use non original stuff
Thats incorrect
First most art is compriced by my ither artists thats how inspiration works
Using ai and getting to that point neans it likly is original
Using ai and relying on ai are 2 different things
So its very likly it is original
Using ai and relying on ai are the same thing imo. If at any point in the writing process you feel you absolutely must use ai for anything, even if it's just for inspiration, instead of just doing it the old fashioned way, you're relying on it for a part of your writing to some extent. You're the one writing the prompt, yes, but you're still depending on what it answers back to you to help make your final decisions for the final piece.
Professional writers write to earn money. Publishers publish to make money. Agents, editors: in it for the money.
Current copyright law only protects works created by HUMANS. there is no money to be made publishing a book that is not protected by copyright.
MONEY, people.
everyone agreed that Falade’s book was sensational. No dispute about the quality of the book. $2 million dollars. that’s a lot of clams.
but publishers aren’t going to publish a book that has copyright issues, where possible AI use puts the copyright under a cloud of illegitimacy.
Ai is only a tool and current copywrite only req 50% human
The book did not have copyright issues they backed down over biases full stop
There was no just reason for the back out ai is not the issue as to get to that point he had to put in enough effort
his agents had a bias against earning their $300,000 commission on the competitive auction sale?
Wtf are you on about
Ai is only a tool a person made a book boot was goid the person making the deal had bias just cause ai was assumed to be used
Its that simple dont try to strawman ir gaslight it wont work
i still don’t understand your point. his agents were the ones who brokered the $2 million deal. the deal didn’t happen out of thin air. the agents planned the auction, they invested a lot of work in making the book deal happen. that’s what agents do. they make the best deal for their clients — and they work on commission. a $2 million dollar book deal means $300,000 commission for themselves.
His agents were trying to get it yes
It was nit the agents that nade it fall threw so idk what you are on about the book deal failed cause of biases
i still don’t understand your point. his agents were the ones who brokered the $2 million deal. the deal didn’t happen out of thin air. the agents planned the auction, they invested a lot of work in making the book deal happen. that’s what agents do. they make the best deal for their clients — and they work on commission. a $2 million dollar book deal means $300,000 commission for themselves. that’s an enormous pay day.
they loved the book and they believed in the book. and they worked hard to orchestrate an amazing book deal of $2 million dollars.
they knew he was black when they orchestrated the book auction. the publishers didn’t pull out of the deal. the agents themselves pulled out of the deal after they became aware of issues involving AI use. they had previously vetted the book with regard to use of AI, and initially they were satisfied. then after the book deal, they learned some new information, which made them cancel the deal. they are not disclosing what the issues were. so no one is in position to argue about what they learned — we simply don’t know what they learned that made them withdraw their support.
we can only assume that what they learned was something substantive, because it caused them to sacrifice their $300,000 commission.
he has retained an attorney and he plans yo sue. if he sues, then all the details will get hashed out in the trial. it doesn’t make sense to jump to conclusions without knowing the details. but right now the only ones who know the details are the author and the agents. the publisher with the winning bid doesn’t even know the details.
the larger point is that we are in a very choppy period for publishing. AI has upended all the rules, norms and expectations that have been in place for centuries. there is no reliable way to determine AI use, except by the author’s acknowledgment and admission. so in this case, it is very hard to imagine what the agents learned that caused them to withdraw their support for the book. the writer is not admitting to AI use and he has hired a lawyer. so it this story is not over yet.
Who gives a crap? If it’s a good book (entertaining and worth the time to read it), why does it matter?
The guy who produced it had to put some effort into it, so why shouldn’t he be able to sell it for whatever people are willing to pay?
I just don’t get what all the fuss is about.
the thing is, i doubt any of these books are gonna be fun, considering all the repetitive, predictable and boring stuff ai writes. i think allowing this kind of stuff in the market may result in a massive amount of crap that keeps growing, possibly overshadowing real, good stories from really talented authors whose life are built around their writing
A good friend who loves WWII crime novels and has read tons of them says the more recent ones are starting to sound alike.
But that's simply not true. Depending on how the parameters are structured, how the prompt is worded, how the contextualized engineering is done, the temperature, the p-value, all the other attributes and parameters and components associated to it produce completely different outputs.
I didn't read the article because I've heard a lot lately here on substack that the tools that are coming out to verify AI use or not in articles is having a lot of false positives. That might not be the case here because you would think if the agents pulled it and that costs them a lot of money there might be more to it. But relying on a computer program to verify if something else is a AI is a little crazy and symbolizes all that wrong with rapid AI adaptation.
I write for a living where sometimes AI is accepted and sometimes it isn't. I'll say this: I've been flagged for AI when I didn't use it and have skated past these flagging systems when I did. Using technology to verify technology use is not at all helpful or accurate.
Exactly my point
Since the beginning of time, all ideas — written or spoken — are derived or borrowed from somewhere. Using AI to help develop a story is no different than using a thesaurus to discover different words. If we’re accepting AI chat bots as necessary in society, then we need to accept that AI will help write stories, ideas, history, etc… from now on.
Solomon said it best, there's nothing new under the Sun.
So true!
All ideas? That's a totalizing view that denies human creativity.
Creativity is a shared trait-we create from our experiences with each other.
But everyone creates something different. To say everything is borrowed means that nothing is original.
You’re right, we create differently from what we’re exposed to. Ideas from other sources are heard, then reacted to. Nothing is original. Everything in existence is created from something else.
I don't see the issue with using AI for research and copy editing. My novels are set in 1993 and in order to accurately describe settings and scenes, LLMs are extremely helpful in jogging my memory.
This is dumb. The book is either good or it isn’t. It will sell or it won’t (the whole reason it was “worth” the $2M deal)
Writing was once controlled by the few who possessed the education, time, and institutional access to produce it. Then came the printing press, public education, the typewriter, desktop publishing, the internet, and self-publishing. Each innovation broke a form of scarcity, threatened established gatekeepers, and inspired warnings that inferior work would flood the culture. Some of today’s resistance to AI belongs to that same pattern. But AI also introduces something genuinely new: the printing press democratized the distribution of human writing; AI democratizes the production of plausible language itself. The real question may not be whether AI writing is legitimate, but who gains the power the old gatekeepers lose and who ultimately owns the new “printing press.” The more useful historical lesson is that democratization and exploitation can occur simultaneously. The printing press liberated knowledge, but printers became new gatekeepers. The internet allowed anyone to publish, but platforms eventually controlled visibility. AI may enable people who were excluded by education, disability, language, money, or professional networks to express complex ideas but a handful of corporations may own the systems through which they do it.
You nailed it. The human owns the complex, non-trivial themes and ideas. They now have AI to help express themselves. They should proudly declare that they used AI to create a manuscript that would otherwise not exist. Then it's for the readers to judge the value of the content.
Spot on! I've been struggling to articulate that as clearly as you have - so many thanks.
Truthfully, his has become a witch hunt. I'm not defending AI, God, no. But the fact that we have to dumb down our creations to not be able to be "detected" as AI is insanity. Just bcs someone wrote something good, doesn't mean they used AI. Just because someone can write fast, doesn't mean they used AI. They were robbed for something alleged that wasn't even proven. Ridiculous.
Absolutely agree with you.
Maybe I'll vex these witch hunters by proposing this:
I've decided not to use AI.
Instead I will move to a country where they don't have any child labour laws and teach kids under the age of 5 to read all the great works of fiction and non-fiction. And then - paying them nothing but scraps and sesame seeds, I'll get them to write my next big book to be sold for $200 million (upfront of course) and no one will be the wiser. No one will even suggest to visit me at my hideaway in international law grey zone or look at my compound to see these children writing and typing on refurbished MacBooks and Surface tablets. They won't even bother with sitting with me and just talking about my work to see if I actually have the intelligence to create such a work. They just pay me... and then ... later... discover I was using kids... illegally... to write my $200million manuscript.
This AI witch-hunt has become its own sub-genre in the long line of people who are unhappy their mediocre writers and think other mediocre writers are achieving great results by simply typing a prompt "make me an award winning novel, chop chop."
In truth, they are envious they didn't think of it sooner. But now... they're unhappy it's not actually possible since they'll be discovered at some point. So really... this whole accusation is coming from people who can't see themselves writing so brilliantly they'll be the next [insert writing hero] unless they got the boosted abilities afforded by some AIs. BUT... they do not want to rely on AI. Circular argument for people unhappy with the challenges of real writing and creative output.
Has Proust been run through Panagram? Hemingway? Dostoyevsky? Have any of you written a book with the help of AI? How different it is from having an editor? or a white board? Do we have to scale back to chisels and stone to be authentic? Do only those who can afford a human editor or have the right connections or pedigree merit access to formal publishing? How much did the editor, the sales team, the publisher, the designer, the cover art affect the final output? How many drafts were there? Why is that acceptable? Shouldn't the prose be pure and untouched? Was research done? Did the author look at a royalty free picture of a castle to help describe their castle? Did they listen to music to generate a vibe while recording notes? Should Taylor Swift be collecting royalties?!
Spot on.
“I also would not hire a ghostwriter to write my work. I became a writer because I like writing. The appeal is in the act itself.”
My favourite few sentences. It’s why I read. It’s why I write. It’s why I sell books. It’s the appeal of the written word, the gift of an author/writer to me.
My writing may never be published, it may always remain a “draft”, but it is all mine. I try to imagine Austen or Tolkien using AI … yeah, nah. AI may have its uses, it may be a tool for checking etc. what it isn’t is living and breathing and experiencing life. What it isn’t is human.
Thanks for this Stack, it’s brilliant.