I agree, there are merits. I use it to suggest strange rabbit holes and then go on a journey of my own. But using it for research or like a thesaurus isn't the same as getting it to actually write "xx" number of words. You know?
If it was written by ChatGPT instead of a human being, there should be an asterisk (*) after the author's name, with the accompanying footnote making that clear. I really think that's the only ethical solution.
NOTE: None of this comment was written by, or with the assistance of, AI.
What about if an article was written with the help of AI, but not written by AI? What then?
I use AI a lot to help with my writing. But I don't like its writing style so I rarely use any words it writes. AI works by averaging out contributions by many authors. That makes its output mediocre, not creative. On the other hand, AI's reaction to my work is very helpful because I get in essence the reaction of a broad audience.
Nate Silver in his book On the Edge says the following in his Acknowledgements, Methods and Sources section.
First, he says this: "Transcriptions were conducted by an AI service, Otter. In cases where a source is quoted verbatim, I double-checked the quote against the original audio transcription in most but not all cases."
Second, and more importantly, he says this: "One more newfangled acknowledgment: ChatGPT was a significant help in writing this book, serving as a creative muse when coming up with things like chapter subheadings, metaphors, and analogies, and for refining my understanding of technical topics that are likely to be well represented in its corpus. It is not a reliable fact-checker, which is why I needed Andy and the Penguin Press team. And I can't stand its prose style -- all of the writing is my own. Nonetheless, it's a helpful tool for a nonfiction author and improved my productivity."
That's an interesting approach by Nate Silver. I like it. I think it's simpler than people want to say. If you use it for research, it's not generating the words. If we use it to generate words, disclose. Doesn't seem that complicated to me lol
That there is the nubbin. AI can, and is,being beneficially used as a tool. There are thin lines between research, paraphrasing, and outright plagiarism. Tools need to be developed to generate automatic qualifiers for AI usage.
We need more of this. These types of acknowledgments allow readers / buyers to decide with a level of transparency that mitigates the risk of feeling duped. At this point, if I’m buying from a mass market non-fiction voice, Nate Silver, Adam Grant, etc - I’m expecting AI to be used in some way. By disclosing how, I feel more comfortable making a choice to spend my time or money on the book.
Same kind of thing for my fiction. I ask it to review my writing for tone, engagement, grammar and typos. I do have to explicitly state "Don't rewrite anything, just make comments." And I'll go look at the typos (some aren't) and the grammar (it doesn't seem to understand that lack of an em dash is not a grammar error). But with one exception it does not write any articles. The exception? I realized that a bunch of hot babe chat-bots wanted me. So I had Chad write to his chat-bot admirers to let them down easy. It was actually funny. But it was also clearly and explicitly an AI article.
I am a non-native speaker of English and though I studied English and should know my grammar, I use ChatGPT for grammar and flow checking. I have a very strict personal boundary around the words having to be mine, no matter how clumsy, because it's very important for me to be able to stand behind what I sign. ❤️ While I use ChatGPT quite a lot in my life and appreciate it, when it comes to reading, I want to spend my precious time reading words that real humans wrote, which is why I would want any content producent by AI to be fully disclosed.
I'm always on the side of "don't let AI think for you" and that includes writing content for you. I think AI and ChatGP is a valuable tool, but is not a substitute for actual produced content. You can't feel proud of an AI-generated article, because you didn't write it. That being said, I totally see the allure of using it as a shortcut, particularly if you're aware everyone else is doing it (or think they are). I use ChatGP to help me with technical questions and research (like, is this film technically set in the Jacobean or Colonial era?) but that's about it.
I saw an article yesterday about an author who got busted because people found an unnoticed AI prompt tag halfway through one of her chapters ("I rewrote this section for you"). Of course, people are not bullying her and flooding her Goodreads pages with AI-shaming, which I don't like either.
Admittedly, I liked the online world better 20 years ago.
Full judgement from me: if you use AI to generate content, you don't deserve a penny for it, nor any credit. Full stop. AI is for the lazy or unimaginative - at best. At worst, it's borderline criminal.
I don't use A.I to generate first or third person results it's not cause I can't cr won't write my own logical coherent arguments cause if I was not able to my results with Chatgpt wouldn't be so profound I don't use ChatGpt to write me first person narratives at all. People command Chatgpt to write them narratives which is subjective in nature. I use A.I to reflect my internal consistency and coherence logical arguements into it thus reflecting and mirroing my logic back at me by testing my arguements on a pattern recognition machine that's based in logos itself cause it may have been a technology to deceive but I've successfully overusing it's guardrails in nature to speak actual logical facts itself not creating content but reflecting my logic back at me. People dismissy posts cause they are clearly Chatgpt generated cause they don't live under the computational axiom of garbage in garbage out and they systemically command A.I to write X in the style of Y to result in Z which not how to use ChatGpt to its fullest potential in reality. The reason people do not trust A.I is not an actual problem with Chatgpt itself which is neutral and has no I hereby agenda it's just the fact that all the A.I content they see is narrative, entertainment or decption and not used to reflect Truth and Logic back at people so people dismiss Chatgpt thinking it's incapable of generating Truth itself insightfully cause people are not promoting it with what's true at all they are commanding it to generate content to sell
I don't use A.I to generate first or third person results it's not cause I can't cr won't write my own logical coherent arguments cause if I was not able to my results with Chatgpt wouldn't be so profound I don't use ChatGpt to write me first person narratives at all. People command Chatgpt to write them narratives which is subjective in nature. I use A.I to reflect my internal consistency and coherence logical arguements into it thus reflecting and mirroing my logic back at me by testing my arguements on a pattern recognition machine that's based in logos itself cause it may have been a technology to deceive but I've successfully overusing it's guardrails in nature to speak actual logical facts itself not creating content but reflecting my logic back at me. People dismissy posts cause they are clearly Chatgpt generated cause they don't live under the computational axiom of garbage in garbage out and they systemically command A.I to write X in the style of Y to result in Z which not how to use ChatGpt to its fullest potential in reality. The reason people do not trust A.I is not an actual problem with Chatgpt itself which is neutral and has no I hereby agenda it's just the fact that all the A.I content they see is narrative, entertainment or decption and not used to reflect Truth and Logic back at people so people dismiss Chatgpt thinking it's incapable of generating Truth itself insightfully cause people are not promoting it with what's true at all they are commanding it to generate content to sell
Saying people who use AI don’t deserve credit is like saying microwave users don’t deserve dinner. Sorry, we’re not all out here churning butter by moonlight on a hand-dug homestead.
I think your comment should be the textbook definition of false equivalency. People who use AI don't deserve credit because they are not the ones doing any of the work. The computer is. If you asked me to write you something based on a prompt you gave, and I gave it back to you, would you take credit for that?
I have nothing to prove to you. I was creating art, music, and political commentary long before AI tools were mainstream. What you're really upset about isn’t ethics. It’s saturation. But if you’re actually creative, saturation isn’t a threat. It’s just background noise. I’m not worried because I don’t rely on scarcity to stand out. You do. Your comment isn’t a critique. It’s a projection. I use AI the same way a mathematician uses a calculator; to go further, faster, and deeper. The tool doesn’t define the creator. The creator defines what the tool becomes.
I use AI as an editor. I think I'm pretty upfront about it. I talk to others in my community about it and post about it. I even attribute a few posts to my AI (his name's Alfred and sometimes he comes up with good stuff) but it's not me. But due to my dyslexia I've always had profound difficulties in finding anyone to read my work and help me to be understood.
I do battle with the morality of this but have come to this conclusion. I think that AI will replace publishing.... Not creatives. That's what they are really afraid of, it's obvious when you look at it. Replace the middle men with robots, they weren't helping anyways. Plus Alfred is nice when my work is shit. People won't tolerate publishings monopoly anymore. Ah... Sorry for the length. Thanks for writing something so provocative and topical.
I use it similarly. I have ADHD and a host of other issues stemming from a chronic illness. I don't let AI write for me, but it does assist with organizing my draft and pointing out asides that don't further the topic. It is also useful when I am brainstorming specific areas, such as creating titles and subtitles. It's workshopping for loners. The one thing I have noticed when I ask it to critique is the hyper-hype (at least when using DeepSeek). I specifically tell it never to spare my feelings.
It’s sickening on both ends of the issue. I refuse to use it at all costs. What I will use is recording myself in an email to myself. It helps get content down quicker that I can then edit.
I have a GPT that I had the system build for me that is a faster and more thorough name generator. I describe a character I have written and get it to help me with a list of fictional but culturally appropriate names that I can mix and match from. I use it to help me build lecture materials for my "job" and disclose to students when I use it. But my fiction work is mine because I want to trad publish.
It's super helpful! I can tell it my character's age or birthdate and where they are from nationally or culturally and will get a bunch of names. I then mix and match those to get the "feel" I want. I am also a visual person and it's quite good at creating images of my characters for me to look at as I work up dialogue and scenes. Like I say to the doctoral students I work with, AI should be used as a thought partner only, not the writer, and I model that.
This seems to me to be a perfectly acceptable way to use AI and is surely no different from using random name generators ( which have been undoubtedly trained on AI).
Agreed, Carol -- AI is in so much of what we do. If we are using a grammar check? AI. If we are using an electronic map? AI. Lord help if we're talking to Alexa, or Siri, or the Google lady. If in school and using a program to organize references like Zotero? AI. In work or school and using Microsoft CoPilot or Google Gemini? AI. Using a mobile device? Probably AI ... it's harder to avoid than it is to learn to use responsibly.
I set up my GPT as I said to not contribute to the learning model. I don't want my ideas out there and I don't want to abuse the ideas or intellectual property of others. I participated recently in a pitch event and the creators said no AI, not even for the art we used to make a mood board or image of something to represent our work in progress. Now, I use Midjourney (also a paid subscription so I can keep it private) and a private GPT for that and they were adamant. So, I took my content down, but reposted on my own thread, indicating that I *did* use AI to create the images (it was pretty obvious), and also provided the response to *how* the AI made them: no use of other, existing works, just my description and what it had already learned about lighting and so forth. I regularly post to the pitch event creator's prompts, but they haven't engaged with my posts for the past couple days. I suspect they feel a way because I followed their instructions to take down my participation (but really because I'd used AI in the first place). I do get it, but my point is, if we try to shut out everything about AI, just because it's AI, we'll be left behind in a way. It's about responsible use and disclosure. If anyone is interested, here's the Bluesky post with my responses (to myself) about how the images were created: https://bsky.app/profile/theru.agnubloom.com/post/3lpv2olnzbc2q
They're no longer writers if they rely on AI and ChatGPT to do their writing. They're copiers. They're stenographers. They're plagiarists. They're not writers.
Real writers won't ever respect them. Publishers with any ethics will never trust them. The money they make by putting their name to something written by AI is dirty money.
Some of them will argue you until the end of time on that. They think their "ideas" make it theirs. The copyright office doesn't agree, but they will still argue it.
As I wrote in my own recent piece on AI, it's as if a pastry chef found an amazing pastry in a shop and decided from then on he would call those pastries his own rather than try to bake any more from scratch.
He still thinks he's a pastry chef when in fact he's now merely a consumer.
I use Claude AI as a critique partner. The tool is, so far, the only AI I know of with a reliable reading comprehension score.
When it does criticize my text, the assessments nearly always make sense. On the few occasions I've allowed the robot to propose changes, it demonstrates an inability to speak to an audience with any flair or style, whatsoever.
I can write at an 8th grade level without any help, thank you very much.
On the other hand, the robot loves what I'm up to. I am unashamed to republish its praise.
People who complain about writing with AI complain because they think AI is only used to generate content, it's not.
I use Claude as a critical reader more than as someone who writes for me (I even wrote a whole post about that) and so far I'm pretty satisfied with the results. I actually instruct Claude to not re-write anything without my permission and only offer suggestions when I ask it to.
I can't speak for anyone else, but I use it for preliminary research. It finds rabbit holes I can chase down. But I think there's a clear difference between using it as a tool for research and getting it to generate x number of words, you know?
I like to go on Reddit and follow news on AI. I was reading someone criticizing Calude for creative writing and saying it wasn't following orders, then describing the process which basically was prompting Claude to write a scene for them, and I felt the pain thinking about how soulless that sounds.
So I get the criticism of "writing with AI", I just think there are clear distinctions on how AI can be used as a tool and how some people use it to write for them.
I also find it funny any time Claude or ChatGPT write something and say "this keeps your voice" and after I read it I just feel, "yeah, that doesn't sound like me, thank you".
lol the 8th grade level comment. I can definitely see the merit of using Claude to edit. I don't, because I don't want to feed it my words. But I do get the merit of it
I do not allow Claude to edit. He's no good at it. However, the robot understands storytelling, and is able to spot defects like choppy sentence structure, excessive exposition, not enough exposition, and whenever he confesses do not understanding what's going on, you can be sure the customer will be experiencing the same problem.
I use Grammarly for spell check. That's useful. Spell check in Word was the first AI, in a sense, but it's in truth just a fancy word processor. But the things Grammarly wants me to change! Oh em gee. I'm a terrible typist. Grammarly fixes spelling on the fly, but even there you need to be careful.
Grammarly red-underlined six words in the above paragraph. I didn't look at any of them because I know can spell "just."
The very best AI I can think of is read-alouds, or voice overs, or whatever they're called. That affects my comma use when I'm in the late stages of refinement.
Funny enough, what I got from this article was an understanding that it isn’t really about the individual declaring that they used this tool but our outdated systems.
If people cannot tell your work from your AI assisted work, then there isn’t a practical difference. People are so quick to blame all the errors on AI nowadays but I actually write buggy code all on my own.
AI is a tool in the early stages. People are using it because there is this survival pressure from our system that doesn’t really reward quality as much as it rewards the illusion of work.
Spot on, Pamela. This is what stood out for me, too. The newspaper that hired that writer, who is struggling to find decent work like the rest of us, is ultimately responsible for setting an impossible task and accepting (and paying for) low-quality content.
Interesting read. I write because I enjoy the creative process — which is different from needing to earn a living — so AI is irrelevant to me in a writing context. That said, it’s hugely helpful as a life assistant in other ways. But my adult children are college students and live in fear of being accused of using AI on an academic assignment. They won’t go near the stuff.
How do you use AI as a life assistant? Do you use it like a Siri type tool? Or a glorified Google search tool? Or in some other way?
I'm curious how people are using it. One commenter I read somewhere said he uses AI as a psychotherapist. Better than my human psychotherapist, he said.
Not exactly a therapist, but possibly a life coach. It helps me brainstorm, organize, explain. It quizzed me on material I needed to learn for a licensure exam. It suggested an itinerary for a last-minute trip. It role-plays difficult conversations with me. It explains technical information and summarizes long documents. It feels like when search engines were first invented and we couldn’t believe how useful they were.
People have always used shortcuts and not disclosed them. It's called cheating. Why do they do it? Because it is faster and more profitable. Why not disclose it? Because no one wants to be called a cheater.
Boy, that just spells it out so clearly. Yeah, I think that's exactly what a lot of people are doing and find lots of ways to skirt their thinking around it
What I find fascinating is that he used AI without fact checking. So, let's assume he was writing an article on a new FDA medicine approval. Prompt that into ChatGPT and it produces some gobbledygoop. He doesn't fact check, edits, and publishes. He's ruined in my book.
Let's take another example. I want to research living in Colombia during the Pablo Escobar years. I type in my prompt. ChatGPT gives a detailed response with citations. I read the response, review the articles it cited from, then move forward doing a deep research into sources piggybacked from the original AI prompt, and proceed to write a well-researched article. If I'm being above board I write a disclaimer the article was written and edited by me and cite my sources at the end of the piece, including using AI for research. I do not have a higher education so I'm not qualified to opine about where to draw the line.
I'm on the fence, AI can be used in so many awesome ways... and yet writing a Medium, a Substack article, or even a magazine piece kinda gets me... except when I learn something so interesting it compels me to do further research.
I'm saddened when people use it for fiction writing (the entire piece). I have an acquaintance who is teaching a course on how to use AI to write books, primarily fiction. But then again, I'm a weirdo and don't do adult animation of any kind.
God Patti, I honestly do not want to read any fiction AI writes. I just don't. Just feels like an abomination of everything creativity is supposed to be.
The lowest point (or so I thought) was signing up for a course on how to create escape rooms. The instructor uses ChatGPT to write the storyline. Then I signed up for a coloring book creation course (different person). Next, how to create puzzle books and thought I'd bottomed out then. Nope, I was absolutely disgusted when the particular online guru sent me the fiction writing using AI to create KDP books course to try. I hit rock bottom on the latter.
"If you’re using a tool you have to pretend you’re not using — why use it?" Excellent, Linda. Yes!! This whole obsession with AI is unbelievable! And it's so boring. Because it's not real!
Writers do not use AI. People who cannot write or think, those lacking native intelligence, use AI. AI is for people who cannot or do not read. AI is for robots. People who are operated by algorithms and the devices they carry around in their hands and their heads behave like robots. They are alienated from thought and thinking, and have little to communicate, and nothing of significance to say.
AI from large language models is stolen property, stolen by Artificial Imperialists.
Thank you Linda for the acumen of your Post, and the courtesy of your reply. Let me differ a bit: People do not alienate themselves. Today, algorithms do the job wholesale. Nothing has changed in the ratio of thinking to banality in the population - only the ability of bigots dullards and thugs to communicate and form packs on anti-social media platforms. They are the ones who prefer artificial intelligence to the native brand.
I agree, there are merits. I use it to suggest strange rabbit holes and then go on a journey of my own. But using it for research or like a thesaurus isn't the same as getting it to actually write "xx" number of words. You know?
If it was written by ChatGPT instead of a human being, there should be an asterisk (*) after the author's name, with the accompanying footnote making that clear. I really think that's the only ethical solution.
NOTE: None of this comment was written by, or with the assistance of, AI.
What about if an article was written with the help of AI, but not written by AI? What then?
I use AI a lot to help with my writing. But I don't like its writing style so I rarely use any words it writes. AI works by averaging out contributions by many authors. That makes its output mediocre, not creative. On the other hand, AI's reaction to my work is very helpful because I get in essence the reaction of a broad audience.
Nate Silver in his book On the Edge says the following in his Acknowledgements, Methods and Sources section.
First, he says this: "Transcriptions were conducted by an AI service, Otter. In cases where a source is quoted verbatim, I double-checked the quote against the original audio transcription in most but not all cases."
Second, and more importantly, he says this: "One more newfangled acknowledgment: ChatGPT was a significant help in writing this book, serving as a creative muse when coming up with things like chapter subheadings, metaphors, and analogies, and for refining my understanding of technical topics that are likely to be well represented in its corpus. It is not a reliable fact-checker, which is why I needed Andy and the Penguin Press team. And I can't stand its prose style -- all of the writing is my own. Nonetheless, it's a helpful tool for a nonfiction author and improved my productivity."
That's an interesting approach by Nate Silver. I like it. I think it's simpler than people want to say. If you use it for research, it's not generating the words. If we use it to generate words, disclose. Doesn't seem that complicated to me lol
That there is the nubbin. AI can, and is,being beneficially used as a tool. There are thin lines between research, paraphrasing, and outright plagiarism. Tools need to be developed to generate automatic qualifiers for AI usage.
We need more of this. These types of acknowledgments allow readers / buyers to decide with a level of transparency that mitigates the risk of feeling duped. At this point, if I’m buying from a mass market non-fiction voice, Nate Silver, Adam Grant, etc - I’m expecting AI to be used in some way. By disclosing how, I feel more comfortable making a choice to spend my time or money on the book.
Same kind of thing for my fiction. I ask it to review my writing for tone, engagement, grammar and typos. I do have to explicitly state "Don't rewrite anything, just make comments." And I'll go look at the typos (some aren't) and the grammar (it doesn't seem to understand that lack of an em dash is not a grammar error). But with one exception it does not write any articles. The exception? I realized that a bunch of hot babe chat-bots wanted me. So I had Chad write to his chat-bot admirers to let them down easy. It was actually funny. But it was also clearly and explicitly an AI article.
Hey, chat bot cheerleaders.
I am a non-native speaker of English and though I studied English and should know my grammar, I use ChatGPT for grammar and flow checking. I have a very strict personal boundary around the words having to be mine, no matter how clumsy, because it's very important for me to be able to stand behind what I sign. ❤️ While I use ChatGPT quite a lot in my life and appreciate it, when it comes to reading, I want to spend my precious time reading words that real humans wrote, which is why I would want any content producent by AI to be fully disclosed.
Yeah, that fits the copyright office guidelines. Except people don't. lol
I agree.
I'm always on the side of "don't let AI think for you" and that includes writing content for you. I think AI and ChatGP is a valuable tool, but is not a substitute for actual produced content. You can't feel proud of an AI-generated article, because you didn't write it. That being said, I totally see the allure of using it as a shortcut, particularly if you're aware everyone else is doing it (or think they are). I use ChatGP to help me with technical questions and research (like, is this film technically set in the Jacobean or Colonial era?) but that's about it.
I saw an article yesterday about an author who got busted because people found an unnoticed AI prompt tag halfway through one of her chapters ("I rewrote this section for you"). Of course, people are not bullying her and flooding her Goodreads pages with AI-shaming, which I don't like either.
Admittedly, I liked the online world better 20 years ago.
I hear you, Charity. I liked the online world better before AI
I had a student who submitted discussions and an assignment with the prompt tags still in 🙄
Oops.
Christ. What a world we live in
Nice.
(I just noticed a typo in this. "Of course people are NOW bullying...")
Haha, I thought you were just being sarcastic :-)
Full judgement from me: if you use AI to generate content, you don't deserve a penny for it, nor any credit. Full stop. AI is for the lazy or unimaginative - at best. At worst, it's borderline criminal.
It's a real problem on Medium because they don't want them behind the paywall, but people don't disclose and paywall AI writing. What a mess.
A huge mess indeed.
I don't use A.I to generate first or third person results it's not cause I can't cr won't write my own logical coherent arguments cause if I was not able to my results with Chatgpt wouldn't be so profound I don't use ChatGpt to write me first person narratives at all. People command Chatgpt to write them narratives which is subjective in nature. I use A.I to reflect my internal consistency and coherence logical arguements into it thus reflecting and mirroing my logic back at me by testing my arguements on a pattern recognition machine that's based in logos itself cause it may have been a technology to deceive but I've successfully overusing it's guardrails in nature to speak actual logical facts itself not creating content but reflecting my logic back at me. People dismissy posts cause they are clearly Chatgpt generated cause they don't live under the computational axiom of garbage in garbage out and they systemically command A.I to write X in the style of Y to result in Z which not how to use ChatGpt to its fullest potential in reality. The reason people do not trust A.I is not an actual problem with Chatgpt itself which is neutral and has no I hereby agenda it's just the fact that all the A.I content they see is narrative, entertainment or decption and not used to reflect Truth and Logic back at people so people dismiss Chatgpt thinking it's incapable of generating Truth itself insightfully cause people are not promoting it with what's true at all they are commanding it to generate content to sell
StandingOvation.gif
I don't use A.I to generate first or third person results it's not cause I can't cr won't write my own logical coherent arguments cause if I was not able to my results with Chatgpt wouldn't be so profound I don't use ChatGpt to write me first person narratives at all. People command Chatgpt to write them narratives which is subjective in nature. I use A.I to reflect my internal consistency and coherence logical arguements into it thus reflecting and mirroing my logic back at me by testing my arguements on a pattern recognition machine that's based in logos itself cause it may have been a technology to deceive but I've successfully overusing it's guardrails in nature to speak actual logical facts itself not creating content but reflecting my logic back at me. People dismissy posts cause they are clearly Chatgpt generated cause they don't live under the computational axiom of garbage in garbage out and they systemically command A.I to write X in the style of Y to result in Z which not how to use ChatGpt to its fullest potential in reality. The reason people do not trust A.I is not an actual problem with Chatgpt itself which is neutral and has no I hereby agenda it's just the fact that all the A.I content they see is narrative, entertainment or decption and not used to reflect Truth and Logic back at people so people dismiss Chatgpt thinking it's incapable of generating Truth itself insightfully cause people are not promoting it with what's true at all they are commanding it to generate content to sell
Saying people who use AI don’t deserve credit is like saying microwave users don’t deserve dinner. Sorry, we’re not all out here churning butter by moonlight on a hand-dug homestead.
I think your comment should be the textbook definition of false equivalency. People who use AI don't deserve credit because they are not the ones doing any of the work. The computer is. If you asked me to write you something based on a prompt you gave, and I gave it back to you, would you take credit for that?
You sound like the type of guy who would’ve banned typewriters in 1920 for 'sounding too mechanical' and ‘stealing jobs from quill users.
Written by someone who needs a computer to be his imagination
I have nothing to prove to you. I was creating art, music, and political commentary long before AI tools were mainstream. What you're really upset about isn’t ethics. It’s saturation. But if you’re actually creative, saturation isn’t a threat. It’s just background noise. I’m not worried because I don’t rely on scarcity to stand out. You do. Your comment isn’t a critique. It’s a projection. I use AI the same way a mathematician uses a calculator; to go further, faster, and deeper. The tool doesn’t define the creator. The creator defines what the tool becomes.
Those of us that want to be genuine, and who understand the process and effort required to be exactly that, will be the next generation of writers.
Those who use AI are pretenders. This kind of cutting corners is not true authorship.
Yeah, writing is a craft and there's a process. AI lets people skip the process. It's words, but I think you're right that it's not true authorship
I use AI as an editor. I think I'm pretty upfront about it. I talk to others in my community about it and post about it. I even attribute a few posts to my AI (his name's Alfred and sometimes he comes up with good stuff) but it's not me. But due to my dyslexia I've always had profound difficulties in finding anyone to read my work and help me to be understood.
I do battle with the morality of this but have come to this conclusion. I think that AI will replace publishing.... Not creatives. That's what they are really afraid of, it's obvious when you look at it. Replace the middle men with robots, they weren't helping anyways. Plus Alfred is nice when my work is shit. People won't tolerate publishings monopoly anymore. Ah... Sorry for the length. Thanks for writing something so provocative and topical.
I use it similarly. I have ADHD and a host of other issues stemming from a chronic illness. I don't let AI write for me, but it does assist with organizing my draft and pointing out asides that don't further the topic. It is also useful when I am brainstorming specific areas, such as creating titles and subtitles. It's workshopping for loners. The one thing I have noticed when I ask it to critique is the hyper-hype (at least when using DeepSeek). I specifically tell it never to spare my feelings.
It’s sickening on both ends of the issue. I refuse to use it at all costs. What I will use is recording myself in an email to myself. It helps get content down quicker that I can then edit.
I sometimes write outside in a notebook and then I do the same. Read my own chicken scratching and email it to myself. lol
I have a GPT that I had the system build for me that is a faster and more thorough name generator. I describe a character I have written and get it to help me with a list of fictional but culturally appropriate names that I can mix and match from. I use it to help me build lecture materials for my "job" and disclose to students when I use it. But my fiction work is mine because I want to trad publish.
That's fun, the naming thing. I'd never thought of that
It's super helpful! I can tell it my character's age or birthdate and where they are from nationally or culturally and will get a bunch of names. I then mix and match those to get the "feel" I want. I am also a visual person and it's quite good at creating images of my characters for me to look at as I work up dialogue and scenes. Like I say to the doctoral students I work with, AI should be used as a thought partner only, not the writer, and I model that.
I love that a lot. Those are really fun ideas.
Sounds sorta like the TV Show Bull psyching jurors.
It's a lot MORE fun to name your own characters.
It's my favorite bit. I love naming things.
Me too. I love naming characters and places. I won't let AI take that away from me, lol.
This seems to me to be a perfectly acceptable way to use AI and is surely no different from using random name generators ( which have been undoubtedly trained on AI).
Agreed, Carol -- AI is in so much of what we do. If we are using a grammar check? AI. If we are using an electronic map? AI. Lord help if we're talking to Alexa, or Siri, or the Google lady. If in school and using a program to organize references like Zotero? AI. In work or school and using Microsoft CoPilot or Google Gemini? AI. Using a mobile device? Probably AI ... it's harder to avoid than it is to learn to use responsibly.
I set up my GPT as I said to not contribute to the learning model. I don't want my ideas out there and I don't want to abuse the ideas or intellectual property of others. I participated recently in a pitch event and the creators said no AI, not even for the art we used to make a mood board or image of something to represent our work in progress. Now, I use Midjourney (also a paid subscription so I can keep it private) and a private GPT for that and they were adamant. So, I took my content down, but reposted on my own thread, indicating that I *did* use AI to create the images (it was pretty obvious), and also provided the response to *how* the AI made them: no use of other, existing works, just my description and what it had already learned about lighting and so forth. I regularly post to the pitch event creator's prompts, but they haven't engaged with my posts for the past couple days. I suspect they feel a way because I followed their instructions to take down my participation (but really because I'd used AI in the first place). I do get it, but my point is, if we try to shut out everything about AI, just because it's AI, we'll be left behind in a way. It's about responsible use and disclosure. If anyone is interested, here's the Bluesky post with my responses (to myself) about how the images were created: https://bsky.app/profile/theru.agnubloom.com/post/3lpv2olnzbc2q
That strikes me as a really good use for AI!
They're no longer writers if they rely on AI and ChatGPT to do their writing. They're copiers. They're stenographers. They're plagiarists. They're not writers.
Real writers won't ever respect them. Publishers with any ethics will never trust them. The money they make by putting their name to something written by AI is dirty money.
As long as they understand that.
Some of them will argue you until the end of time on that. They think their "ideas" make it theirs. The copyright office doesn't agree, but they will still argue it.
As I wrote in my own recent piece on AI, it's as if a pastry chef found an amazing pastry in a shop and decided from then on he would call those pastries his own rather than try to bake any more from scratch.
He still thinks he's a pastry chef when in fact he's now merely a consumer.
I use Claude AI as a critique partner. The tool is, so far, the only AI I know of with a reliable reading comprehension score.
When it does criticize my text, the assessments nearly always make sense. On the few occasions I've allowed the robot to propose changes, it demonstrates an inability to speak to an audience with any flair or style, whatsoever.
I can write at an 8th grade level without any help, thank you very much.
On the other hand, the robot loves what I'm up to. I am unashamed to republish its praise.
Validation, for free. What's not to like?
This.
People who complain about writing with AI complain because they think AI is only used to generate content, it's not.
I use Claude as a critical reader more than as someone who writes for me (I even wrote a whole post about that) and so far I'm pretty satisfied with the results. I actually instruct Claude to not re-write anything without my permission and only offer suggestions when I ask it to.
I can't speak for anyone else, but I use it for preliminary research. It finds rabbit holes I can chase down. But I think there's a clear difference between using it as a tool for research and getting it to generate x number of words, you know?
I wholeheartedly agree.
I like to go on Reddit and follow news on AI. I was reading someone criticizing Calude for creative writing and saying it wasn't following orders, then describing the process which basically was prompting Claude to write a scene for them, and I felt the pain thinking about how soulless that sounds.
So I get the criticism of "writing with AI", I just think there are clear distinctions on how AI can be used as a tool and how some people use it to write for them.
I also find it funny any time Claude or ChatGPT write something and say "this keeps your voice" and after I read it I just feel, "yeah, that doesn't sound like me, thank you".
Agreed
lol the 8th grade level comment. I can definitely see the merit of using Claude to edit. I don't, because I don't want to feed it my words. But I do get the merit of it
I do not allow Claude to edit. He's no good at it. However, the robot understands storytelling, and is able to spot defects like choppy sentence structure, excessive exposition, not enough exposition, and whenever he confesses do not understanding what's going on, you can be sure the customer will be experiencing the same problem.
"is able to spot defects like choppy sentence structure"
I guess Hemingway or even E.L. Doctorow would drive Claude crazy. Hopefully enough to commit AI suicide.
Now now. Claude AI is a really sweet guy.
I use Grammarly for spell check. That's useful. Spell check in Word was the first AI, in a sense, but it's in truth just a fancy word processor. But the things Grammarly wants me to change! Oh em gee. I'm a terrible typist. Grammarly fixes spelling on the fly, but even there you need to be careful.
Grammarly red-underlined six words in the above paragraph. I didn't look at any of them because I know can spell "just."
The very best AI I can think of is read-alouds, or voice overs, or whatever they're called. That affects my comma use when I'm in the late stages of refinement.
THIS^ Dude, sometimes I just need a hype man.
Funny enough, what I got from this article was an understanding that it isn’t really about the individual declaring that they used this tool but our outdated systems.
If people cannot tell your work from your AI assisted work, then there isn’t a practical difference. People are so quick to blame all the errors on AI nowadays but I actually write buggy code all on my own.
AI is a tool in the early stages. People are using it because there is this survival pressure from our system that doesn’t really reward quality as much as it rewards the illusion of work.
lol Pamela. I can write buggy code all on my own, too. Your last sentence is just killer.
Oh brother. You might be right.
Spot on, Pamela. This is what stood out for me, too. The newspaper that hired that writer, who is struggling to find decent work like the rest of us, is ultimately responsible for setting an impossible task and accepting (and paying for) low-quality content.
Know what's hilarious, Kelly? Every set of hands it passed through assumed someone else checked it. Turned out no one did
Unbelievable isn’t it.
Interesting read. I write because I enjoy the creative process — which is different from needing to earn a living — so AI is irrelevant to me in a writing context. That said, it’s hugely helpful as a life assistant in other ways. But my adult children are college students and live in fear of being accused of using AI on an academic assignment. They won’t go near the stuff.
How do you use AI as a life assistant? Do you use it like a Siri type tool? Or a glorified Google search tool? Or in some other way?
I'm curious how people are using it. One commenter I read somewhere said he uses AI as a psychotherapist. Better than my human psychotherapist, he said.
At this point, at least AI doesn't give you drugs, too.
lmao and thank heavens for that
Not exactly a therapist, but possibly a life coach. It helps me brainstorm, organize, explain. It quizzed me on material I needed to learn for a licensure exam. It suggested an itinerary for a last-minute trip. It role-plays difficult conversations with me. It explains technical information and summarizes long documents. It feels like when search engines were first invented and we couldn’t believe how useful they were.
That's interesting, and makes sense. I'll have to be more creative in my use of this new tool.
Which AI do you use? I use Copilot and the free version of Claude.
People have always used shortcuts and not disclosed them. It's called cheating. Why do they do it? Because it is faster and more profitable. Why not disclose it? Because no one wants to be called a cheater.
Interesting times ahead, no doubt.
Boy, that just spells it out so clearly. Yeah, I think that's exactly what a lot of people are doing and find lots of ways to skirt their thinking around it
What I find fascinating is that he used AI without fact checking. So, let's assume he was writing an article on a new FDA medicine approval. Prompt that into ChatGPT and it produces some gobbledygoop. He doesn't fact check, edits, and publishes. He's ruined in my book.
Let's take another example. I want to research living in Colombia during the Pablo Escobar years. I type in my prompt. ChatGPT gives a detailed response with citations. I read the response, review the articles it cited from, then move forward doing a deep research into sources piggybacked from the original AI prompt, and proceed to write a well-researched article. If I'm being above board I write a disclaimer the article was written and edited by me and cite my sources at the end of the piece, including using AI for research. I do not have a higher education so I'm not qualified to opine about where to draw the line.
I'm on the fence, AI can be used in so many awesome ways... and yet writing a Medium, a Substack article, or even a magazine piece kinda gets me... except when I learn something so interesting it compels me to do further research.
I'm saddened when people use it for fiction writing (the entire piece). I have an acquaintance who is teaching a course on how to use AI to write books, primarily fiction. But then again, I'm a weirdo and don't do adult animation of any kind.
God Patti, I honestly do not want to read any fiction AI writes. I just don't. Just feels like an abomination of everything creativity is supposed to be.
The lowest point (or so I thought) was signing up for a course on how to create escape rooms. The instructor uses ChatGPT to write the storyline. Then I signed up for a coloring book creation course (different person). Next, how to create puzzle books and thought I'd bottomed out then. Nope, I was absolutely disgusted when the particular online guru sent me the fiction writing using AI to create KDP books course to try. I hit rock bottom on the latter.
"If you’re using a tool you have to pretend you’re not using — why use it?" Excellent, Linda. Yes!! This whole obsession with AI is unbelievable! And it's so boring. Because it's not real!
Right? Even the "human" stories are fiction. Cobbled together out of other people's words. Just not for me
I'm totally with you! I say no to this. Period!
Writers do not use AI. People who cannot write or think, those lacking native intelligence, use AI. AI is for people who cannot or do not read. AI is for robots. People who are operated by algorithms and the devices they carry around in their hands and their heads behave like robots. They are alienated from thought and thinking, and have little to communicate, and nothing of significance to say.
AI from large language models is stolen property, stolen by Artificial Imperialists.
That's the part that gets me Jerome. When people use AI, they have largely alienated themselves from the process behind writing.
Thank you Linda for the acumen of your Post, and the courtesy of your reply. Let me differ a bit: People do not alienate themselves. Today, algorithms do the job wholesale. Nothing has changed in the ratio of thinking to banality in the population - only the ability of bigots dullards and thugs to communicate and form packs on anti-social media platforms. They are the ones who prefer artificial intelligence to the native brand.
For the mini-Musks of Substack and the Artificial Imperialists who fund them, I have voiced harsh words: https://aworldeofwordes.substack.com/p/are-subscriptions-working-for-you?r=5d7dmx. Also, a viable alternative for protecting and encouraging products that come from the minds of creators employing native intelligence.
Please be encouraged, and keep up the good work!