The Lazy Decision That Cost an Author Everything - Shy Girl Controversy

Shy Girl, a horror novel was pulled by a major publisher after an AI detector flagged it

In this episode of Honest Brew, we unpack the Shy Girl controversy and why this story is bigger than one book. It raises questions about trust, reputation, lazy decision making, and what happens when probability gets treated like proof.

We also talk about ethical AI use, where discernment still matters, and why creators should care even if they never plan to publish a book

Link to read more about it: 

Link to read it: 
https://www.nytimes.com/2026/03/19/books/shy-girl-book-ai.html

A candid conversation between three seasoned business women who've been in the trenches of entrepreneurship. We bridge the gap between the glamorous just market and sell advice and the reality of what it takes to build a sustainable business. While most business content focuses on marketing, branding, OR operations in isolation, we bring all three worlds together. Because your brand culture needs to live in every system you create, your operations need to support your brand promise, and your marketing needs the infrastructure to deliver on what it sells.

We're here for the solopreneurs ready to grow beyond themselves, the partnership survivors rebuilding stronger, and anyone tired of business advice that treats branding, marketing, and operations as separate planets when they're part of one ecosystem.

HOSTS

Cheale Villa, Visual Caffeine, visualcaffeine.com / Monique Johnson, MoJo Design, ...

Transcript

SPEAKER_00 0:04

Poor novel just got pulled from one of the biggest publishers in the world. Not because it had bad reviews, not because it had poor sales, but because an AI detector said it was AI written. The author says she didn't do it, her editor did, without telling her. The publisher panicked, the book was canceled overnight, and somewhere out there, a 1990s novel is getting flagged for AI by technology that didn't even exist when it was written. So today we need to talk about something every single one of us in business are navigating right now. And you might think, why are we talking about books? I don't want to publish a book. Well, these conversations are important to all of us because it is a greater conversation about AI content and the detection that seems to be policing it. So with that said, ladies, we have to brew this so deeply. And I have so many, let's just put devil's advocate to put to this one. And the first one I'm gonna throw out to you guys is here comes, here comes. If people love the book, which is why it was published in the first place, why did this even happen?

SPEAKER_01 1:20

When I looked into this, I did see that some avid readers were upset with the writing of it. Like they thought it was poorly written. And I feel like that was a discussion that was happening before this whole AI thing happened. And I'll make sure that the article that I read it from, because it was from a woman who used to work in publishing. So her perspective on this was very interesting. But what I really want to touch on is I love reading. And I will be reading my beautiful fictional book, and I will come across an M-dash. And my brain will be like AI when it's like, no, they are using proper grammar. It's just AI who likes to use it in every other sentence. Right. So it does make me almost kind of nervous thinking of kids in middle school and high school using AI and it like messing with like what is actual correct grammar and spelling and punctuation because we get so mad when we see the M-dash or you're not this, you're that sentence structure. So I know that's a bit off of a tangent from a book, but I feel like that applies to overall copywriting because we are in business and we all will do some type of copywriting within our business.

SPEAKER_02 2:27

Definitely gets a bad rap for sure. How M dashes people it is interesting how every single time I see an M-dash, I think it's AI. And it's it's like that that poor piece of grammar really got a bad rap to AI. It's like, you know, but it does make you think like you need to adjust your writing so that you don't seem like you're AI. Like that's a crazy thought. It's sad.

SPEAKER_00 2:56

I you guys are so getting me like because one, I love end dashes. I've loved n dashes before AI ever existed. Two, um, I think there is a certain style choice that grammar, there's grammar and then there's style choices of writing. Also, where I want to go on you basically what I'm what I'm hearing of what you guys are saying is it's kind of making me laugh as you're talking about grammar. Now let's go back, not even 10 years. Let's go to the blog days, right? Okay, so the blog days kind of between blogging and social media. Um, can we just say grammar what? What was grammar? Huh? What are you talking about? Because grammar went completely out the window. So it's really hilarious to me for people to start getting on their high horse about grammar. I'm just being realistic here. We have been communicating with horrible grammar for years. AI, if anything, AI came back to, hey guys, we need to go back to grammar. I'm not doing it well, but hey.

SPEAKER_02 4:02

Yeah, but don't you don't you think that um grammar is the easiest part of language for AI to grasp onto?

SPEAKER_00 4:12

Well, I agree, but what's what's okay? I'm gonna put another note on that. You can you could take what you got out of AI, put it in grammarly, and then it's corrected. People are just lazier than ever.

SPEAKER_01 4:22

Yes, fair enough. I also think what's interesting with the AI detectors, though, is that AI doesn't know. It has to be trained to know. So I feel like at this point, you could throw, say, a Sarah J. Mass book into it, and it'll come back being like, this is not by a person, this is by a tool, when like that's not true. So I feel like we need to be wise with how we're using it, especially I because my fiance came across this um in colleges, how they're like very against AI and they're using detectors, and students have been like, no, I wrote this, but because the AI thing says it's 90% AI, they fail. So it's like, I think there's two things happening here is AI is getting like almost too much credit. Cause we have to remember we're the ones who's training it, it doesn't know. And we're also relying on it too much. Like that professor should have just read the paper instead of being like, I'm gonna throw it through AI to detect it, give it me a summary, whatever. It's like we're becoming lazier no matter what realm we're looking at.

SPEAKER_00 5:25

Back to this story. The editor said that they put some parts in AI to edit it, but an editor does isn't supposed to be rewriting, they're supposed to respect the work, right? They're supposed to be just editing it. So we're talking about a really bad editor or a really lazy editor anyway. Yeah, I don't reversely am just saying, based on the information that's out there, that seems like not even correct way of editing. That's not even editing. Also, I want to bring to attention is the fact that this was a self-published book that was so popular that a publisher was drooling and said, we're gonna re-release this under our our publish, our publishing house. So I think it's important to recognize that because that was one reason I said in the beginning, if it's so popular, what difference does it make? I'm not saying I agree with any of it. I'm saying there is a point of devil's advocacy that's like, well, if I'm enjoying this, and I'll give you a different example with the Iran war. Last thing I want to really mention here, but Iran is putting out these Lego movies that are made with AI, and they are brilliant. They are hilarious, they are they really lighten it, but it's funny because they're coming from another. It's another country's propaganda. Let's just be realistic. But they are really entertaining, they're made by AI. Yes, I would rather an animator is spending the time to do that, but humans obviously can't put out that that quickly, right? The point is, is that I think there also is a conversation in here of a balance because AI is here, it will be used. And what does balance look like? And the other side, of course, of the book thing is well, then everybody's gonna start doing AI books, and they're gonna, you know, we're gonna fill our bookshelves with all these AI books, and we don't want that either, obviously. And that's the part that my heart breaks because I am a book reader, love old literature, you know, with all of that said. What do you guys hear in that?

SPEAKER_02 7:36

Yeah, I I think that context is a huge part of this whole story. It changed the meaning of the whole story when you just said that it was self-published, and then the publisher picked it up. And so it's like just knowing the full story and the full context behind each one of these pieces really changes the final meaning of it, you know? And so I just think that we need to take a step back and be like, where did we originally see this? What is the intention of it? How am I perceiving it? And how is the whole story being affected by AI or how what is AI's part in this whole thing? So, you know, for me, it's I think it's really important to just really look at where it was published and the audience it was intended for, and the entertainment value of it. All these things are are factors. There's a million variables. And for, you know, for one person to put it through their lens only, I think it's kind of an unfair judgment.

SPEAKER_01 8:40

There's a lot of layers to the conversation because the article I read, it also I feel like points to issues in just the industry of publishing and editing and like the contracts that people are signing, saying to it like to attest that like AI is not a part of it, or long large language models. And it's just like policies need to change. And also like the way that a lot of publishers take on books now is based on what is their popularity on social media. Which it's like that's kind of like a red flag. Cause again, yes, we want to make sales, we want to make money, that's how we live. But it's also like, are you just gonna put out a crappy book because they did really well on Reddit or a fanfic site? Or are we gonna stick to actual like proper writing, proper, like going through the proper process so people and authors get the I was gonna say reward, but that's not the recognition that they they deserve. And that the author, the editor, the publisher, they are all respected within their industries and they feel safe to flag if an author is plagiarizing or using AI and vice versa.

SPEAKER_00 9:44

I love that you both said what you did because I love the context conversation and I love the what I'm hearing from you, Sarah, is the integrity conversation.

SPEAKER_02 9:54

Yes.

SPEAKER_00 9:55

And we're losing, we've lost integrity. Um, and when we are picking things because they're popular, nothing against the people who enjoy it, but people also enjoy fan fiction and all sorts of things that are not well-written pieces of literature.

SPEAKER_02 10:11

Yeah.

SPEAKER_00 10:12

So there's a standard, obviously. And with that said, it I'm I'm I'm remembering a story, and this was, of course, before AI. Well, AI probably existed in the back rooms. I don't know if you guys remember that Australian woman that claimed that she had brain cancer and her diet of juices and good eating. Yes. Yes. There's a whole documentary on her. I watched it. Yes. It was a very good one, by the way. But what you guys are talking about brings me present to that. She was a complete fraud. Totally fraud. Not only did she get an app deal with Apple, she got a huge book deal. Yep. She was exposed at book, like she was exposed around the time that her book was published. Yep. Now, with that said, I remember when this woman was popular, it was around the time that I had a um a crunchy vegan blog, and she had good recipes. There was good things. She was a liar, she was a fraud, and she stole from people. I'm not saying that. Totally recognize that. But being going back in time, there was good things that she was putting out. There was good recipes. I don't think we need to demonize people 100%. I think that we need to also pay attention to everything and the whole picture. And I think that our that our we have a very reactionary society where we go from one extreme to another and we actually aren't like looking at a bigger picture. I am not saying that this girl should have continued to be able to do what she's doing. Not saying that at all. But I think we are quick to demonize people. And what my understanding of this story is the only thing that has been admitted to is there's some portions of this book that were the editor had put into AI. The whole book wasn't, if we believe the author, the whole book was not written by AI. But instantly, it's demonized, instantly, it's canceled, instantly. All of these things happen because we have this reactionary society. All of this happened because we had a detection tool that did this. And this is something that I want to lead us to. One AI detection company admitted its tool is simply unfamiliar with older writing styles, which is why it flags public domain works, old journals, classic literature as potential AI generated. We are talking about ready for this? The Constitution. The Bible people, Jane Austen, all flagged. Um, so the AI detectors are essentially AI detecting itself. And here's where Mr. Freddie Via's wisdom comes in. He says, remember, AI only knows what it knows because of what we fed it.

SPEAKER_01 13:05

Say it for the people in the back. Yes, like I just for the people for the people. For real, like to be doing it to ourselves. Yes. I think for me, this kind of makes me think of I don't know if it was Amazon or a different company that let go a bunch of people because they're like, oh, AI can do it. And then they were like, oh shit, no, they can't do it. I just feel like we're trying to be too efficient too fast, and we're not asking good questions. And that's why we're running into scenarios like this, especially with this detection thing. That should have been a no-brainer to take Shakespeare, Jane Austen, all of these different ways of writing, because I think of my bookshelf, each of those authors writes so differently. Are some of them complete garbage? And I question how they got a book deal. Yes. And there's other books that I'm like, I will lay my life on the line to get the next book. And that also comes down to personal preference. So I just feel like with how rapidly companies are trying to put robots out there, do AI out there, it's like we need to take through a grain of salt and actually let it work out the kinks because it's just going to backfire in a month, in a year, 10 years from now, it's going to backfire if we don't take proper precautions.

SPEAKER_02 14:26

100% agree with that. And I also wanted to bring up this topic of the publisher and them feeling like their reputation is suddenly at risk if they don't pull it and there's this big scandal that's going on, then what do people say about them? You know, and so there's a whole nother world of just like the publisher industry. And I mean, I don't know, Sarah, because you read a lot of books and you probably know more about this topic than I do. But what is like, I mean, how do these, because I feel like the publishing industry is is it's kind of old school, but in a good way, how do they like manage this whole AI? How do they navigate this AI world at this point?

SPEAKER_01 15:13

And the article, it was having like the really strict clauses and contracts being like you have to attest that this is original work and that there's no AI generated images or text at all. And a lot of people don't think as think of AI as plagiarizing, which is kind of like it kind of is because it's pulling that information from other places. Yes, because it is taught by a human, it doesn't just know. And so I just feel like we're all trying to get on this AI bandwagon. We see how it can make us more efficient, in which being too efficient can hurt us, and that's coming from an operations person who loves that shit. And I think also, too, we are afraid of being canceled because of how quickly people can destroy someone's reputation over one thing.

SPEAKER_00 16:03

Right. And I think there's a few really beautiful points there. One is that whole fear of cancellation, the other is what you talked about with different writers. Like you read different writers. Like I'm a huge Shakespeare person. Some people can't stand Shakespeare because they can't understand it. And that brings me to one of my points that I wanted to bring. And being a neurodivergent person, this one really hits home because I put things in on a detectors that I have written and it gets it gets hit. Neurodivergent writers, people with autism, ADHD, I personally have ADHD, or and non-native English speakers are flagged at higher rates than native speakers because their natural patterns resemble what detectors have been trained to call AI. These AI detectors are curbing, you know, curbing opportunities for people if we are getting so legalistic to use them as the holy grail. And like I said before, my husband's point of like it only knows what we feed it. Well, who is AI to detect AI? Yeah. That might be that some people might think that's a dumb thing to say, but I mean, really think about that.

SPEAKER_02 17:20

No, it's not.

SPEAKER_00 17:21

So there was one writer that um uploaded her own teenage journals. This is written like before large language models were even existed, once again. And the detection flagged it was 75% likely AI generated. Forgive me. 70% likely AI generated. These are written journals, people. So I think that there is, if we go back to anything new that has entered our society, we even had this with those, I can't remember what they were called. People were, you know, the the art, the digital art that people are spending tons of money on. Oh yeah, those monkey, the digital image of a and it was like certified. And my husband actually really got into those things and stuff, but I remember how I felt when those came in. And I I think that it's kind of a similar feeling for me at least, but I feel like it might resonate with you ladies, which is that there is an inherent originality, creativity, integrity. And I know there's a few other words in there that I will not say at this time, but that those things are absent of. And I think that that's what the detection truly is. There is something absent here. And I will say with one word, it's a human soul. I agree.

SPEAKER_02 18:42

Yeah.

SPEAKER_00 18:43

You can write like AI, but your soul is present, your passion is actually present in it. But with that said, it's like you can have preferences of things and you can have certain patterns of way that you speak that inherently bring someone to a place of like a detector may go, I, but a human with a soul can hear the human in there.

SPEAKER_01 19:06

Yes. Cause you can tell the difference in a social media caption of someone who just had AI do it and they slapped it versus someone who may have used AI to get a first draft or an outline, but then went in and edit it and then rewrote it. Because I'm gonna be honest, in my marketing, that's kind of how I use AI because I will overthink my ideas. I struggle with the creativity side of my business. So this helps me take my word vomit, structure it into something where I'm like, oh, I see the framework, I see how I'm leading someone to booking a call with me. Now let me go in and rewrite it so it sounds like me. And there might even be mistakes in there, but like, y'all, I am not the best at grammar. So if there's a spelling, we're not gonna die over it.

SPEAKER_02 19:54

I've actually just recently, I was on LinkedIn and a um content writer said that sometimes like it's like we're getting to the point where you're putting mistakes in there on purpose to like prove that you're human.

SPEAKER_00 20:10

That's so wild to me.

SPEAKER_02 20:11

I know, I never thought of that before, and I was like, wow, is that what we're coming to at this point? You know, it's just I just feel like we've jumped on the bandwagon so quickly on this journey, and it's like the wild west right now. Like there are no rules and regulations. Like, I'd like to fast forward. I'm not saying um there's no value in AI. Uh I I use it daily myself in the ways that Sarah just described. But I also think that if we fast forwarded a year or two, maybe even more, there would be more clarity on best ways of using it or what it's good at, what it's not good at, or some kind of definitions or rules around it so that we can get a grip on how every how we are working together with AI as humans.

SPEAKER_00 20:59

Yeah. And and like I said before, you know, you who is watching this might wonder why are we talking about something publishing when we're talking about solar printers and small business people? Well, because there is ways that AI is really supporting us. Me being an ADHD creative person, AI supports me in just helping me be a little more organized in my thinking. Does that mean I have it replace me? Absolutely not. There, you know, and that's not what it's about, and that's not what it should be used for. But these conversations are important because your company using AI or not using AI, do you really want to see your company canceled because someone decided to flag your company for detecting on AI detectors? Like, do we really want to be policed by AI for AI? It makes no sense. I don't want AI being my police, I want a human being my police. So, with that said, I think that the other thing that we noted to that though is laziness. All of this is. Is happening because we don't want to want to spend the time to do the things that we are supposed to be as humans doing. And AI is really just meant to help us be more efficient time-wise because we have high demands on time. And our viewer uh is a small business owner. You are a small business owner that is listening. Um, and we know that you need to have that time back from you to do the most important things. So let's make sure that the demonization is not happening from using it the way it should be ethically used, and also keeping the conversation open for creators, true creative people that have a soul in what they're creating, that they have protection at the end of the day. So I'm gonna wrap up with this quote AI learned how to write from humans. So when a detector flags your content as AI, it might just mean you write well. The problem is treating probability as proof. With that said, bless sips.

SPEAKER_01 23:26

It's going to be a blend. And that blend is dependent on what you value most in your company. And there will be stories out there like Shy Girl, and I'm sure there's many others that we haven't even touched on. And it's about staying true to your values, using your discernment. And also, like, it doesn't hurt to keep the receipts if you have some random random person come out after you because you'll have proof.

SPEAKER_00 23:50

Thank you for your brilliance. And I just want to leave you all with the human feeling behind words still matter. And that's what we ultimately are searching for.