Is someone you know using AI to write?

[gettingby]

An acquaintance wanted to talk to me about writing, but then when I met up with her went on about how AI is such a great "tool" for writers. She explained how she's using it to get ideas, but then she would pick and choose the ideas she liked and rewrite everything herself. She believes by doing this she does not have to tell anyone she used AI since she was not using anything word for word.

I tried to explain the best I could that AI is not a "tool" for writing, and that by using AI to even assist with writing is a big no-no for writers and publishers. Her response was super defensive and she seemed to dig her heels in about how great AI is. It got a little weird, and I tried to get back to writing talk in general, but she kept bringing up AI like she was trying to convince me that all writers will eventually use AI and that it's the wave of the future. At the same time she said no one would know so there really wasn't any problem in her mind.

Then there was her argument that all ideas come from somewhere and that her using AI was the same as wherever my ideas came from because everything comes from somewhere. And then she said there is nothing new in writing so it really was fine to use AI, just as fine as googling something or reading all those books before I wrote my own. She seemed to really believe it was the same thing. I didn't have the energy to go back and forth with her so I ended our coffee meeting as soon as I could without it seeming rude. But maybe I should have been a little rude. She was the one who wanted to meet with me in the first place.

This is a friend of a friend who knows I'm a writer. That's why I met up with her. And the weirdest part of everything is that she wanted me to read some of her writing and wanted to send me not just her writing but what she put into AI from her original idea, what AI gave her, and then her rewriting of it. She said if I saw what she was doing it would change my mind. I tried to tell her I really couldn't get into all that. In the end, the easiest way to get out of there was to give her my email address.

I have no intention of ever looking at her stuff. I also didn't have the energy to fight about AI. I'm just going to ignore her. If my friend we have in common asks, I was planning on telling her I've just got too much going on. Maybe I should tell her that her friend is not a real writer because she's using AI, and thinking that's okay will get nowhere. But not everyone cares that much about writing or understands where AI (LLM) came from and how it was created and how it really has no place in writing. I don't think my friend would care that much, and her friend certainly doesn't care at all.

What would you say to a so-called writer who uses AI? I think some people have convinced themselves that LLMs are a good thing. What a sad and scary thing for writing and writers.

 

[Unimportant]

What would you say to a so-called writer who uses AI?

If they wish to engage with me in a conversation about it, I say all the things that I and dozens of others have already said in all the threads in this section of the forum. Feel free to read them at your leisure.

If they don't wish to engage with me in a conversation about it, I don't say anything to them because why would I?

 

[Woollybear]

Usually I say that the reason I ride a bicycle is to gain fitness. I don't ride electronic bikes, because doing so would go a ways toward defeating that purpose. Other people ride electric bikes, (but not without getting a little side-eye from some of us within the general population... but it's their choice) and there are legitimate reasons some people ride electric bicycles. But I won't, because when I get on a bike... it is to make my body move the bike. I feel spent and very good afterward.

Same thing with writing. I do it for 'writer fitness.' I personally wouldn't use an AI to create a narrative because my brain is not getting the amount of exercise I want for it when the AI does the bulk of the work.

We grow through effort. Process, not product. Etc.

 

[Paul Lamb]

Another consideration about AI that is often overlooked is that it is essentially plagiarism. It is reading and using text (generally without permission of the original author) and building from that to "create" a new bit of writing.

Many of the journals I submit to require me to affirm that I did not use generative AI to produce any part of my work. (Things like grammar and spell checkers are excepted and accepted.) The contract I just signed for my next novel specifically required me to affirm that I did not use AI. So if this friend did use it and then lied, it would essentially be a breach of contract.

To me, AI takes much of the fun and personal investment out of creative writing. The satisfaction of conceiving a great plot or character or resolution is as much the reason I do this as the seeing it published (and reaping the huge financial rewards).

 

[RichardGarfinkle]

I don't think this person was trying to have a real discussion. So, oddly enough, I don't think it matters that this was about AI.

This kind of meet up sounds like a classic proselyting strategy. The proselytizer arranges a social meeting about a supposed shared interest but really they want you to be converted to their viewpoint. They're not there to listen. They're there to testify to the great new thing that has opened up a whole new world to them etc.

Friend of a friend is a tenuous social distance that can make people uncomfortable. There's a third party kind of being held hostage to the target's good behavior. People don't want their friends to hear that they weren't nice to this other person who just wanted to talk to this writer friend etc.

The thread topic is worth discussing, but I don't think that any good that comes of the discussion will apply to the kind of situation gettingby was in.

 

[Jean P. R. Dubois]

Personally I think AI should stay out of the creative arts domain, like writing and graphic arts. But at this point the cats out of the bag and there's little that can be done to put the toothpaste back in the tube. That said I think any and all AI generated or even assisted art should be mandated to come with a clear label stating as such. Much like how movies, video games, and even food (GMO) are clearly labelled, have a little "AI assisted" label on it so people know where their entertainment came from. I know I would be pretty ticked off if I bought a book only to sense something was off partway in then have to do digging on the author only to find it was mostly a computer writing the book.

"After reading the dictionary, every other book is just a remix."

 

[Cobalt Jade]

Can't speak for writing, but plenty of graphic designers are using AI now.

 

[pyrael]

An acquaintance wanted to talk to me about writing, but then when I met up with her went on about how AI is such a great "tool" for writers. She explained how she's using it to get ideas, but then she would pick and choose the ideas she liked and rewrite everything herself. She believes by doing this she does not have to tell anyone she used AI since she was not using anything word for word.

I tried to explain the best I could that AI is not a "tool" for writing, and that by using AI to even assist with writing is a big no-no for writers and publishers. Her response was super defensive and she seemed to dig her heels in about how great AI is. It got a little weird, and I tried to get back to writing talk in general, but she kept bringing up AI like she was trying to convince me that all writers will eventually use AI and that it's the wave of the future. At the same time she said no one would know so there really wasn't any problem in her mind.

Then there was her argument that all ideas come from somewhere and that her using AI was the same as wherever my ideas came from because everything comes from somewhere. And then she said there is nothing new in writing so it really was fine to use AI, just as fine as googling something or reading all those books before I wrote my own. She seemed to really believe it was the same thing. I didn't have the energy to go back and forth with her so I ended our coffee meeting as soon as I could without it seeming rude. But maybe I should have been a little rude. She was the one who wanted to meet with me in the first place.

This is a friend of a friend who knows I'm a writer. That's why I met up with her. And the weirdest part of everything is that she wanted me to read some of her writing and wanted to send me not just her writing but what she put into AI from her original idea, what AI gave her, and then her rewriting of it. She said if I saw what she was doing it would change my mind. I tried to tell her I really couldn't get into all that. In the end, the easiest way to get out of there was to give her my email address.

I have no intention of ever looking at her stuff. I also didn't have the energy to fight about AI. I'm just going to ignore her. If my friend we have in common asks, I was planning on telling her I've just got too much going on. Maybe I should tell her that her friend is not a real writer because she's using AI, and thinking that's okay will get nowhere. But not everyone cares that much about writing or understands where AI (LLM) came from and how it was created and how it really has no place in writing. I don't think my friend would care that much, and her friend certainly doesn't care at all.

What would you say to a so-called writer who uses AI? I think some people have convinced themselves that LLMs are a good thing. What a sad and scary thing for writing and writers.

I am very much against AI for the actual writing (regardless of media). But I will admit, for research purposes it's quite helpful. And I must also say that while there are programs out there that do spell check, grammar, and even word frequency well. I posted a scene to Grok, and was impressed with how it pointed out not only those things, but also made suggestions on how to better present the text for reader comfort etc. not to mention suggesting similar phrases to keep the writing lively.
I generally don't use AI for much of anything, but for research and grammar checking? I don't see the issue there. It is essentially just a tool at that point. It's when you take a bite of the apple that problems arise. Asking Grok, for example - after discussing a scene and it's improvements - "How would you write it?" or "Can you write that out for me?" not. cool.

 

[lizmonster]

I don't see the issue there.

Well, there's the environmental impact. And the large-scale theft. And the fact that it's fancy autocomplete, and I wouldn't trust a thing it threw at you - including grammar corrections - without checking a non-AI source.

 

[Brigid Barry]

For research it's wrong something like 80% of the time.

For spelling, grammar, and readability, there is software that was not trained on stolen works that does the same thing.

ETA: Grok was a major influence in so many people leaving X because there was no more opting out and it's training on people's posts.

 

[RichardGarfinkle]

I am very much against AI for the actual writing (regardless of media). But I will admit, for research purposes it's quite helpful. And I must also say that while there are programs out there that do spell check, grammar, and even word frequency well. I posted a scene to Grok, and was impressed with how it pointed out not only those things, but also made suggestions on how to better present the text for reader comfort etc. not to mention suggesting similar phrases to keep the writing lively.
I generally don't use AI for much of anything, but for research and grammar checking? I don't see the issue there. It is essentially just a tool at that point. It's when you take a bite of the apple that problems arise. Asking Grok, for example - after discussing a scene and it's improvements - "How would you write it?" or "Can you write that out for me?" not. cool.

How do you use it for research? It has no fact checking capability and most LLMs will make up references. They produce statements that have thee same shape as the results of research, but without the actual research.

 

[dooder]

It's kind of interesting to think about the advancement of technology, and how it affected authors throughout the 20th century.

Up until later into his career, Tolstoy resisted the technological advancement that was the typewriter at the time. He was using pencil and paper!

Cormac Mccarthy started out with a typewriter, and never upgraded to a computer back for the rest of his writing career.

George RR Martin on the other hand has used WordStar 4.0 on DOS, a software released around the 80s, and to this day saves his work on floppy disks.

However, younger authors like Rothfuss, Sanderson, Abercrombie (Relatively young, of course), became adults around the time when Microsoft word became popular, and rendered everything else obsolete. It brought many new techonological functions, like autocorrect and auto-format--none of which were seen before.

It would be meaningless for us to boot up MS97 to write, or to emulate the WordStar in our PC, or switch to a typewriter you bought from the flea market.

Fastforward to right now, we have something like AI which can, if used properly, sometimes act as an assistant to an aspiring author. However, it's still too early to gauge and measure AI's impact on the literature world. I think it's going to make sense in retrospect only. It's a very interesting topic.

 

[CMBright]

The issue, in my opinion, is two fold.

The first is ethical. Both taking copyrighted material from authors and artists that do not want their data scraped and the environmental costs. Using copyrighted material without compensation or permission whole and calling it fair use. Saying they can't possibly create data sets without that wholesale theft of copyrighted material in court.

The other is that AI/LLMs are not fact checkers. They are based on probability. And if they don't have the information, they string together words that sound plausible. Using AI/LLMs for research can be worse than useless.

For me, it is an ethical decision not to participate in wholesale copyright infringment.

 

[lizmonster]

However, younger authors like Rothfuss, Sanderson, Abercrombie (Relatively young, of course), became adults around the time when Microsoft word became popular, and rendered everything else obsolete. It brought many new techonological functions, like autocorrect and auto-format--none of which were seen before.

Well, as someone who's closer to their ages than not - I have no idea when Word started including autocorrect. Every autocorrect function I've seen is actually driven by the OS, and can be turned off there. (Which I do, because for fiction it's worse than useless.)

To my mind, all these technological advancements are nothing more than tool changes. Authors are doing exactly what Tolstoy was doing, except not with pencil and paper. (And please note, there are lots of writers who still draft with pencil and paper. Many of them hang out here.)

Fastforward to right now, we have something like AI which can, if used properly, sometimes act as an assistant to an aspiring author.

There is no "proper" use of AI for an aspring author.

However, it's still too early to gauge and measure AI's impact on the literature world.

Oh, I disagree. It's a scourge, and it will remain a scourge, probably until China ramps up enough for the companies shoving this down users' throats to acknowledge that it doesn't do what they claim it does, and what it does do isn't terribly useful on a practical level.

It's an environment-destroying plagiarism bot. It produces tokens that read like written language. That's an interesting piece of tech, but it doesn't have much use beyond the novelty of it.

 

[pyrael]

How do you use it for research? It has no fact checking capability and most LLMs will make up references. They produce statements that have thee same shape as the results of research, but without the actual research.

I don't know which Ai's you are familiar with, but I have found grok to be pretty good for research. It supplies links to articles so that you can check the sources. I haven't found anything made up from it yet. However, I have received plenty of made up stuff from GPT and the like. I'm not saying that I, or anyone should trust it, just that it's a useful tool in regards to research. And again, just my opinion, but Grok seems far more stable. Still not going to use it as my literary editor or beta reader. Just a decent tool for researching topics and grammar checking.

 

[Brigid Barry]

It's kind of interesting to think about the advancement of technology, and how it affected authors throughout the 20th century.

Up until later into his career, Tolstoy resisted the technological advancement that was the typewriter at the time. He was using pencil and paper!

Cormac Mccarthy started out with a typewriter, and never upgraded to a computer back for the rest of his writing career.

George RR Martin on the other hand has used WordStar 4.0 on DOS, a software released around the 80s, and to this day saves his work on floppy disks.

However, younger authors like Rothfuss, Sanderson, Abercrombie (Relatively young, of course), became adults around the time when Microsoft word became popular, and rendered everything else obsolete. It brought many new techonological functions, like autocorrect and auto-format--none of which were seen before.

It would be meaningless for us to boot up MS97 to write, or to emulate the WordStar in our PC, or switch to a typewriter you bought from the flea market.

Fastforward to right now, we have something like AI which can, if used properly, sometimes act as an assistant to an aspiring author. However, it's still too early to gauge and measure AI's impact on the literature world. I think it's going to make sense in retrospect only. It's a very interesting topic.

Apples and oranges. Aka, the logical fallacy of false equivalency.

 

[lizmonster]

I don't know which Ai's you are familiar with, but I have found grok to be pretty good for research.

I'm glad it's worked for you. Here's someone who found differently. The plural of anecdote is not data.

It supplies links to articles so that you can check the sources. I haven't found anything made up from it yet.

I'm glad you're at least checking its output.

Just a decent tool for researching topics and grammar checking.

What source are you using to double-check its grammar corrections?

 

[dooder]

Well, as someone who's closer to their ages than not - I have no idea when Word started including autocorrect. Every autocorrect function I've seen is actually driven by the OS, and can be turned off there. (Which I do, because for fiction it's worse than useless.)

To my mind, all these technological advancements are nothing more than tool changes. Authors are doing exactly what Tolstoy was doing, except not with pencil and paper. (And please note, there are lots of writers who still draft with pencil and paper. Many of them hang out here.)



There is no "proper" use of AI for an aspring author.



Oh, I disagree. It's a scourge, and it will remain a scourge, probably until China ramps up enough for the companies shoving this down users' throats to acknowledge that it doesn't do what they claim it does, and what it does do isn't terribly useful on a practical level.

It's an environment-destroying plagiarism bot. It produces tokens that read like written language. That's an interesting piece of tech, but it doesn't have much use beyond the novelty of it.

Can’t it be helpful for research? Or with making advanced grammar corrections, especially for non-native speakers? What if an aspiring author struggles to write with perfect grammar, or has trouble spotting errors? Don’t they still deserve a chance to publish their work in a readable format that their works deserves to be? Maybe they simply can’t afford a line editor. I don’t think anyone has the authority to decide what’s “proper” for someone else, as long as it’s not hurting someone.

And of course, I'm not agreeing with all the points I made above. I'm simply trying to raise questions, because I don't think the answers to them are as simple as some people make them out to be.

 

[dooder]

To my mind, all these technological advancements are nothing more than tool changes. Authors are doing exactly what Tolstoy was doing, except not with pencil and paper. (And please note, there are lots of writers who still draft with pencil and paper. Many of them hang out here.)

Of course, but we can make the argument that tools affect our lives more than we realize! I sometimes resorted to pen and paper for brainstorming and outlining, but compared to the 1800s, I'm sure the number of people using pen n paper for writing is very few nowadays.

 

[dooder]

I have no idea when Word started including autocorrect. Every autocorrect function I've seen is actually driven by the OS, and can be turned off there. (Which I do, because for fiction it's worse than useless.)

I believe it was in 1993, which is pretty crazy! I thought it would've been way later!

 

[Woollybear]

When it comes to making progress on personal ideas and projects, I believe these LLMs can help some people some times in some situations. This doesn't address the very real and very substantial ethical concerns of plagiarism to train the things.

Personally, the effort of writing is part of why I write. It feels pointless to me to use a machine to produce the thing--like fast food or mass produced fabric or whatever. The IKEA of writing. IKEA has its place in the world, but I don't shop there. It's a bad metaphor, but these tools are strange and a little icky. Like GMO strawberries that taste vaguely of fish.

 

[lizmonster]

Can’t it be helpful for research? Or with making advanced grammar corrections, especially for non-native speakers?

It's built on large-scale theft from other writers, and it's inaccurate. How is that helpful?

What if an aspiring author struggles to write with perfect grammar, or has trouble spotting errors? Don’t they still deserve a chance to publish their work in a readable format that their works deserves to be?

Okay, let me unpack this a little.

Every writer goes through a learning stage. Good writers never stop learning. If you're aware you have issues with your grammar, keep learning. Keep writing, keep getting feedback, keep reading. There is no shortcut here.

As for deserving a chance to publish their work? No, of course not everybody deserves that.

I enjoy drawing. In the last few years, I've upped my game a lot. I'm also not VanGogh, and I never will be. Do I deserve to be in museums? Do I deserve to have my work auctioned for millions? Of course I don't. Being super nice and having cool ideas doesn't mean you deserve to be paid for your work.

Just because you want to make art doesn't mean you deserve success. You think really wanting something means the public should turn around and pay for it uncritically? Please.

Maybe they simply can’t afford a line editor. I don’t think anyone has the authority to decide what’s “proper” for someone else, as long as it’s not hurting someone.

GenAI is hurting lots of people. It's hurting the entire planet, and jeopardizing the human race as a whole.

And of course, I'm not agreeing with all the points I made above. I'm simply trying to raise questions, because I don't think the answers to them are as simple as some people make them out to be.

Oh, I see. You're just sealioning playing devil's advocate. Thanks for letting us know.

 

[RichardGarfinkle]

Can’t it be helpful for research? Or with making advanced grammar corrections, especially for non-native speakers? What if an aspiring author struggles to write with perfect grammar, or has trouble spotting errors? Don’t they still deserve a chance to publish their work in a readable format that their works deserves to be? Maybe they simply can’t afford a line editor. I don’t think anyone has the authority to decide what’s “proper” for someone else, as long as it’s not hurting someone.

And of course, I'm not agreeing with all the points I made above. I'm simply trying to raise questions, because I don't think the answers to them are as simple as some people make them out to be.

Mod note:
Word of warning: "I'm simply trying to raise questions" is an explicit tool of trolling. If you wish to explore the possible ramifications or usages of something, then make your questions more incisive and illuminating. Don't just ask questions, make the questions work thinking about make them from actual perspectives.

Never post in a way that is indistinguishable from trolling.

 

[pyrael]

I'm glad it's worked for you. Here's someone who found differently. The plural of anecdote is not data.



I'm glad you're at least checking its output.



What source are you using to double-check its grammar corrections?

Interesting article. I can honestly say I've not had that experience. However I'm not usually researching those kinds of things. My queries are usually more abstract and geared toward plausibility of some strange concept for a work of fiction. Like the concept of time being a force similar to light and how that could be plausible.
I always look at the links
The grammar checker in LibreOffice does fine for most of my needs.

 

[neandermagnon]

I would tell the person to have more confidence in their own ability to generate ideas. If they believe they don't have ideas, I'd suggest some techniques to get ideas going, such as brainstorming (think of a topic and write down as many things as you can related to that topic, no holds barred, as crazy ideas as you like - could be a question, scenario or anything else... for example, "write down as many excuses as you can think of that a cheater may tell their partner for why they didn't come home until 3am last night" OR "write down as many ways as you can to cut the grass") - Anyone can come up with ideas with practice and ceasing worrying about the artificial distinction between "good" and "bad" ideas (yes, in the real world not all ideas will actually work, but this is fiction - ideas are what you make of them, and even in the real world, by brainstorming crazy solutions you might hit upon something that could actually work). (BTW this technique can work for getting your character(s) out of plot dead ends - where your character's in a situation and you're like wtf now? How are they going to get out of this situation?)

I mean FFS why use AI when you could have so much more fun not using AI? Creativity is fun.

I'd probably also caution that AI generated ideas are recycled ideas that already exist, i.e. it's going to be generic and derivative. If they want something more original, I'd suggest not using AI.

When it comes to morals/ethics of AI, if you separate the way it was created (i.e. by stealing people's intellectual property and hard work) and the potential uses of it - let's say AI had been created in an ethical way, using it to generate an idea, which you then entirely take on as your own, would not be unethical - it would be no different than basing a story on a published work (or works) and making sure you file off all the serial numbers (like how 50 shades of Grey started off as Twilight fanfiction, but the finished product has nothing that's specific to Twilight in it) It's a difficult question though, because of the fact that these AIs were trained on stolen intellectual property and you can't really get away from. LLMs are permanently tainted by that.

There are other issues with AIs that I've discussed elsewhere, such as that people think they "know" things when all the can do is produce output that sounds like it's from an intelligent human but that is devoid of any ability to fact check. It's a Dunning-Kreuger machine and not to be trusted.