48 - The Human-AI Editing Partnership with Erin Servais of AI For Editors

In this episode of Brave New Bookshelf, we welcome Erin Servais, founder of AI For Editors, who is at the forefront of integrating AI into the editing process. Erin shares her insights on how AI can complement the creative and decision-making strengths of human editors, enhancing efficiency without compromising quality. She discusses her favorite tools, such as NotebookLM for developmental editing and ChatGPT's reasoning models for line editing, while emphasizing the importance of transparency and maintaining oversight in the editing process. Join us as we explore how AI is transforming the publishing industry and learn valuable tips from a pioneering expert. Visit our website https://bravenewbookshelf.com to view the full episode notes, links and apps mentioned in the episode, and the full transcript.

[00:00:00] Welcome to Brave New Bookshelf, a podcast that explores the fascinating intersection of AI and authorship. Join hosts Steph Pajonas and Danica Favorite as they dive into thought provoking discussions, debunk myths, and highlight the transformative role of AI in the publishing industry.

 Steph Pajonas: Hello everyone and welcome back to the Brave New Bookshelf. I'm one of your co-hosts,  Steph Pajonas, CTO of Future Fiction Academy, where we teach authors how to use AI in any part of their process. The area behind me is different than usual. I'm at the beach house for a little while, while one of my kids is traveling in Europe and the other one just got her wisdom teeth out.

So we're enjoying some downtime at the beach convalescing and relaxing and trying to get some sun. It was very rainy here the last few days. I'm hoping for better later on today. And we are recording a few of these episodes during the summer before we release them in the fall. So I'm excited to bank a few of [00:01:00] these and meet our awesome guests and then get to see them again in the fall as we roll through September and into the final quarter of the year. Oh my God. Where's 2025 gone? It's like summer, my summer started like June. My husband's like summer's almost over. I said, stop rushing it. Don't rush it. We don't wanna rush through all these things 'cause that's where all the good times happen and good times have been happening.

I'm excited to see my good friend Danica again, because she's been traveling.  How are you doing today?

 Danica Favorite: I'm good. I'm good. Our last podcast, I was in the throes of this horrible summer cold. It is a week later. I am still recovering. Like Steph was saying, the summer just flies by and you're go, go, go, go, go.

And so I am definitely suffering the effects of that. I was just saying in the preview when we were talking backstage, I don't get sick very often and so when I do, it is a doozy and this [00:02:00] really is like the effects of not resting. And not taking some time for myself to just rest. But and I think that is something, I know that is not the topic of today.

But it has been coming up in our circles lately, so we definitely are gonna be talking about that in the future. For today I'm very excited about today's guest. Actually I'll pause there, because I need to introduce myself. For those of you who don't know me, I'm Danica Favorite. I'm the community manager at Publish Drive, where we help authors at every stage of the journey from formatting their manuscript to finding the right keywords and book descriptions and metadata.

To getting your book published and out to as wide of an audience as possible. And finally, for tracking your sales data and splitting royalties and all of that. And then of course, Steph and the amazing people at the FFA can help you write that book. So between the two of [00:03:00] us, we have everything you need to move forward on your publishing journey.

And what's exciting about today's guest is this is one more piece of the puzzle that people are always asking for, asking about, and I'm really excited we get to deliver this today. Today I have Erin Servais with us and she is an editor. Yes, guys, we have it. Woohoo. Finally, we've got someone to talk about editing and using AI.

And really super excited about this. I have to tell the story, because when Steph and I first sat down to put this podcast together we made like a list of our dream guests. And Erin was on that list and Steph said to me, you've gotta meet Erin. She's great, blah, blah, blah. And I didn't know Erin.

And I was like, okay, cool. So she was on the list of people we'll get to eventually, but I haven't met them, so I have to find a way to meet them. And then I was at the [00:04:00] IBPA conference in Minneapolis and this woman comes up to me and she's hi, I'm Erin, and. We start talking. She's I love your podcast, blah, blah, blah.

And I'm like, oh my gosh. And then she introduced herself and I was like, we've been wanting to have you as a guest. And very pleased that worked out because as you are going to find out from this conversation, Erin is absolutely delightful and super, super excited to have her because, in the AI for Authors group people ask about how to use AI for editing almost every day.

Here we have a professional editor actually coming in to talk about that. I know there's all this fear of can AI edit, is it taking editor's jobs, all of this stuff. And Erin, number one is embracing AI, but number two, she helps other editors use AI in their process, which I think is really [00:05:00] amazing and great.

And so without further ado, I'm gonna turn things over to  Erin Servais, and she is going to tell us all about herself, and then we'll get into today's discussion.

 Erin Servais: Thank you, Danica. I am, I think, officially blushing from that introduction. And I feel honored to be on this list.

I was not aware that I was on that. Thank you for that, for thinking of me for that. My name is  Erin Servais, and as mentioned, I do teach editors how to use Artificial Intelligence. In 2023, I started my company AI For Editors, and I've been going very strong since then. Not taking breaks for rest, like I should and vacation, like I want to, b ecause this has been a topic that people have been really eager [00:06:00] to learn about and driven to learn about. So since summer of 2023, I've been going strong. I have a course that I teach. I now have two levels of it. An introductory AI for editors and advanced AI for editors that anyone can sign up for.

I teach any and every type of editor editing any and every type of content that I could ever imagine from, I haven't checked recently, but the last time I did check it was over 20 countries. I wonder what it is now across the world who've taken my course. So it's been quite popular and which has been a wonderful experience and wonderful to be able to bring this knowledge and very important upskilling and reskilling information to people.

And then [00:07:00] last year I started a spinoff of that company called Servais Strategic Business Solutions, which does custom trainings for companies and organizations, teaching them how to use AI for editing and content creation, content analysis and development that is based in the types of texts that they work on within their groups.

 Steph Pajonas: I love this because you know how we ran across each other. It was on LinkedIn, like of all places, LinkedIn decided one day to serve me your content, which is great because it actually knew that I liked AI stuff and I was looking for AI, uh, other people who use AI like I do. And it served me your content.

And I was like, "Hey, this is great." Because you know what, as authors especially, if you're writing fiction and you're self-published, you actually have to have some sort of editing chops in order to get your book to [00:08:00] a point where it's readable, it's understandable. Like it, even if you can just get it to a point where you can hand it off to an editor and the editor can also read it and understand it, and then make those fine tuning choices and whatnot.

You have to build up those skills as an author. And a lot of authors struggle with that part of the process. They love the creation and the brainstorming and the writing of the prose and everything, and then they finish the book and they're like, oh wait, I have to edit this now?

So we see that a lot.

 Danica Favorite: This is so timely for me and so important because seriously, this is why I stayed trad for as long as I did. Like this is honest Danica talk right here. I hate editing so much, and I was hiring editors and spending a fortune on them, to be honest, and still getting comments back from readers, "This was poorly edited," and I'm like. What the heck? And so it was just easier having trad books, because they just handle that for you. [00:09:00] And then the publisher gets those, what-the-heck-emails. T here you go. Real talk. But I do, I have so many books. We were talking about how I've finished a couple of AI books. They're all sitting on my hard drive guys.

Because the editing part is such a challenge for me, and I love that because. I don't love that for myself because I have these books sitting on my hard drive. But it's important, because like I was saying in the intro, we have people every day in the AI group saying, "Hey, how do you edit? How do you use AI for editing?" And then there's also the concern of what about human editors? And for me, I really love that Erin obviously is a human editor and she's teaching people how to combine that human and AI editing. So let's just get started with the first question about how you're approaching AI in publishing.

Like you said, you saw this summer 2023 and we're like, okay, I've gotta get in on this. As an editor what were your [00:10:00] thoughts on AI coming onto the scene and what you wanted to do about that?

 Erin Servais: The last time I checked I could verify myself as a human. So I'm gonna assume that is, that's still the case.

The question was how I, one more time. How did, what was your, the question,

 Danica Favorite: It was, how are you approaching AI in  policy?

 Erin Servais: How am I approaching AI? With a combination of human and AI, verifiable human and AI, because those are... The human part is absolutely essential to the process. And I see it remaining that way for as long as I can project into the future.

These days it might be a little different than what I would've said before 2022 how far I could predict. But, we are still extremely [00:11:00] integral to the editing process. When verifiably human people use AI for editing that means that they're letting the machines excel at what machines are good at, while reserving humans for what humans are good at.

And I know you both can do incredibly creative and imaginative things with AI in your writing, but thinking about the average user and how they may use AI for editing, that looks something more like leaning on AI to do those rote tedious tasks with editing, maybe some ideation.

Coming up with chapter title ideas, things like that, but humans bring to it this creativity. Our creativity and our decision making, [00:12:00] which is so important. We want to be the one still involved, making those creative decisions. Maybe AI can give us some ideas. But we have to determine which of those ideas is going to be best for the book and best for the author.

So we are making the decision, and we are using our people skills.

 Danica Favorite: Yeah, I really...

 Erin Servais: Which is something AI can't do. So we are connecting with the authors and talking to them and communicating with them, in the way that we can do really well. We know what's going to resonate with them. And we can have that connection, the human to human face-to-face connection that you won't get with an AI, until I suppose we have more humanoid robots, but that's another topic.

 Danica Favorite: It is. But I like what you were saying 'cause I want to go back to that idea of the decision [00:13:00] making and the creative decision making. Because I think that's one of the things Steph and I talk about a lot, is this idea of, as a human, you always have your agency in working with the AI. And ultimately, you know, when Steph talks, Steph talks a lot about validating what the AI is doing, and that's part of that human decision making and that human creativity and human agency. And I really like that relationship part too, because I definitely have a very good relationship with the AI, and it's very loving and very supportive. Like the other day I put something in there. And it came back with something that was just wrong, and I was like, "Hey, are you sure about this?"

And it's oh, let me think about that. And so we had to have this back and forth conversation where I was fact checking the AI. And I thought, with a human, that wouldn't have been the conversation. A human would've not even made that [00:14:00] mistake. Because again, it's that human interaction and connection, but you're so right about that decision making and creativity. Because ultimately we have to, as a human, say, "Was that the right decision by the AI? Do we agree with this? If not, why? And how do we apply that?"

 Steph Pajonas: Editing is a lot of decision making, right? You may have an idea of what is wrong with it. Maybe it's too long. Maybe it's too short. Maybe you're, you've used a lot of passive sentence fragments or whatever it may be, whatever it may be.

You may have an idea about what's wrong with it, and then the AI can come and give a sort of diagnosis of what's going on with your chapter. And then ultimately, you have to make the decision about what you want from that chapter, or whatever you're working on, so that you can get the most out of your work.

And so what I've found from you, Erin, is that you had found that [00:15:00] ChatGPT gave you a leg up in this part of the development process, right? So it's like it's gving you options, and then you're teaching people how to ask for those options and how to decide on what to do with those options and whatnot.

So why don't you talk us through the kinds of stuff that you teach other editors on how to use this technology?

 Erin Servais: I teach people how to use it across the editing workflow and with different types of editing and different types of content. Thinking about the very, the high level developmental editing or content editing, that's probably my favorite step to use AI. And I have been quite enamored lately with Notebook LM and its capabilities. Notebook LM is a...it's marketed as a research assistant. It's by Google, [00:16:00] but it has so many capabilities that are useful for editors, especially data extraction. So I used to be for a long time a developmental editor, mainly for fiction. And I would spend hours upon hours for every single book, creating character lists, detailed character lists, detailed setting lists, detailed timelines, all of these resources that were so time consuming to do on my own, but were very important to help me as I was reading the book and to track different elements and they were good resources for the author as well. But now, with a tool like Notebook LM, you can, with the author's permission, upload their manuscript [00:17:00] to Notebook LM, which has security built into it, so it's kept within that notebook it's called, and it would take... I'm being generous here, and because I think it would actually take less than this, but let's say a minute, for a character list to auto generate a character list or a setting list or a timeline or all of these different things. For nonfiction, you could do glossary. You can do a character arc. Anything that you can think of to have it extract from that manuscript, it can do and put it in a format that will be usable to you.

It is just remarkable. I taught yesterday this lesson I teach about Notebook LM and how to use it for things like this, [00:18:00] and I looked at everyone's faces and they're just awestruck.

 Danica Favorite: I love that.

 Erin Servais: Do you have any questions? And they're like I don't have any questions. I just don't. I just didn't even know it could do that.

And they, it's so unexpected and just beyond what people realize it had the capability to do. T here's so much power there for that. And that's, that is just developmental editing. I teach people how to use it for line editing, like consistency checking. It's really useful for that Notebook in particular is. How many times characters have magically color changing eyes.

And you can stop as you're going along and you're in, let's say, chapter 14, and you're like, wait, didn't Ricky have blue eyes? And you could ask that question to Notebook LM, and it would [00:19:00] then bring up where it said. Ricky had blue eyes or brown eyes, and you can fact check yourself there. And that helps you so much with consistency in a way that would, you could have done before.

I did before, but it was just so much more time consuming.

 Danica Favorite: Yeah I'm really excited about this. And  Notebook LM in particular. Because Notebook LM first came out, I tried to use it for some of that, and it kept hallucinating, which now I know Notebook LM has gotten so much better. So the other day Steph was like, oh, I just used Notebook LM for this and it was so great and blah, blah.

I'm like, oh my gosh, I need to go back to that. And so now hearing you talk about it, I'm like, okay, that's it. I am, I'm signing up for Erin's class. But I do think that is really important for people to realize is that I can already tell you just based on some of the things Steph said, that I'm like, okay, here's where I failed in this process.

So I'm curious, can you give us a few tips to whet our appetite for [00:20:00] some things that people might try in  Notebook LM to give them a little bit more of an edge to get better results? Or what do you find are some of the common mistakes people are making when using  Notebook LM?

 Erin Servais: One thing you mentioned about the hallucinations you noticed with  Notebook LM, which, it surprised me because I have not noticed hallucinations, but I have noticed tiny errors that made sense when you got context. So for instance, I had an experience where I was creating a character list and it said that this particular character had permed brown eyes and I thought. What? That sounds very painful. The nice thing about [00:21:00] Notebook is that you can very easily see where it's getting its information.

It's, it has a little citation number right by each claim. So I hovered over that citation number and a window popped up that showed me exactly where in the manuscript it got that info and, it was in a description of the character and it said, I can't remember exactly what it was, but it talked about how she had her hair was like something like her hair, comma, permed, comma brown eyes, dot, dot, dot.

So when you saw that, you could see how it, it made that mistake of getting permed brown eyes from that. And that's one of those times that it shows, it's important to have that human still involved. Didn't show up somewhere that the character actually did have permed brown [00:22:00] eyes. Her hair. Hair, not eyes,

 Steph Pajonas: Her hair.

 Erin Servais: Yes.

 Steph Pajonas: I was thinking that, I was thinking that one of the things that, that this can highlight is the fact that. If the AI got it wrong, it's quite possible that somebody who was reading that didn't necessarily understand it either, right? In the way that it was constructed. That's that sentence.

And that gives you an idea of what you can go back and maybe, change the sentence structure a little bit to make it a little bit easier to understand. Because then if the AI then understands it, it's more likely that a human will understand it as well.

 Erin Servais: I hadn't thought of that, but that is so true. So true.

 Danica Favorite: Yeah. I like that. And I'm glad you said that because I definitely personally, one of my challenges as a writer is I have really complex sentences that I can see where the AI would be making mistakes like that. And so that's a really good reminder that okay, the [00:23:00] AI, maybe if the AI is getting it wrong, my readers are too, I think that's a really great editing tip. And I love, Erin, what you were saying about, it goes back to the human validation, right? Going back and seeing, okay, this is where the AI got it wrong. Now I know why. It's so cool to me to understand that we aren't asking the AI to do everything for us.

We as humans still have to go back and validate. So why don't you share some more about where we were going with the different line edits and things like that since I know that was where you started to go and we interrupted you to talk about  Notebook LM because I really I am excited about the tool and then playing with it a lot.

So now I want to do even more, but let's go line editing now.

 Erin Servais: I was using Notebook LM for line editing in that it was, I was checking consistency issues with  Notebook LM. So I do like it for the developmental editing, the data [00:24:00] extraction. It's really good for analysis. Line editing is good for the consistency checks, but my, my go-to for line editing are the reasoning models actually. In ChatGPT it's any of the models that have, oh, they, all of these companies seem to have very confusing names for their models.

 Steph Pajonas: I'm always laughing about the fact that like they started with 3.5, then 4, then suddenly there's a 4o nd then there's an o1.

And I was like, oh my God. Can you guys like use real like names? Use like name names? It'd be so much easier for us.

 Erin Servais: Reasoning. One or, yeah, so I like the ChatGPT reasoning models, and that's the ones that have the o in front of the number, not after. So now there's o1, o3, o4 [00:25:00] and they're all pretty darn good for line editing, and you can try giving it a sentence or a paragraph and asking it, you can even be really simple. Please simplify this sentence, and that works really well. Another approach that I like with the reasoning models and line editing is to use an iterative approach. So you'll take a passage not terribly long, we're not talking about even a full chapter necessarily but a page of text, let's say and say, let's reword or take out cliches. Let's eliminate cliches on this pass of it, and then read it through. Identify another issue with it that you would like to adjust whether that's... Oh dear maybe [00:26:00] reducing vague phrasing and then have a prompt where it does that prompt on the entire text. But because the reasoning models do take this more methodical approach, they can do better at staying in their lane and just doing that one thing. Whereas other, the other types of models tend to be, they wanna be really helpful and sometimes that means that they do more than what you asked them to do with editing.

And that's why I prefer the reasoning models for that.

 Steph Pajonas: I prefer them as well. And then something that people need to remember is that AI was made to be helpful but not obedient. So the thing about AI is that it's going to try to be as helpful as possible, and sometimes it goes a little bit too far, just like you were saying, right?

But it's not [00:27:00] obedient, right? So it's going to try to execute the kinds of things that you're asking for, but if it can see a way to be even more helpful by changing a few details here and there that you didn't ask for it will still do it. So you have to keep an eye on it and make sure that it stays in its lane, just like you were saying.

 Danica Favorite: Yeah. I really appreciated a couple things about that. Because number one. Steph, you just hit on something. As I mentioned, I've been sick, and so what I do is I dictate a lot and then I have AI clean it up and et cetera, et cetera. And so being sick, I recently dictated a post from my substack.

And I stuck it in ChatGPT to clean it up. And I, first of all, I did not use a reasoning model. Second of all I was pretty vague in my prompt. 'cause I was sick, I wasn't thinking it through. And so yesterday I reread this post, I was like, oh my gosh, this sounds just like ChatGPT, what the heck? And I looked through it.

I was like, oh my gosh. [00:28:00] It really changed a lot that I didn't ask it to change. I was just like, wait, this was totally me and my prompt fail. And so I think that's always a good lesson in number one, what model are you using? So I love your tip on the reasoning model. But the other thing I wanna call out that I think is really important is so many people expect to stick the whole book into ChatGPT or whatever and get the edits, and you are like, no, I'm not even talking a chapter here.

I'm taking a passage. And I'm saying, okay, this specifically is what I'm asking for to edit in this passage. Because again, like Steph was saying, it's being helpful, so you just want cliches gone. But if you are not specifically saying only, look at the cliches, it's gonna do a bunch of other stuff and change way more than you bargained for. And so I think those are really [00:29:00] two great tips that I hope our listeners are really clueing in on. That idea of, one, pick the reasoning models and two really be specific, and three only do the small chunks. If you're expecting it to do the whole book, that's where it's gonna get hung up.

 Erin Servais: It depends what type of editing you're doing, because there's a different answer to that for each type of editing. If we're talking about doing content analysis, the developmental editing level, when I'm working with  Notebook LM doing content analysis, it's working on the entire manuscript. It can work on 200 plus pages, no problem.

It can work on multiple manuscripts at once. No problem. If you want to verify something's consistent across multiple books, it can do that. But when [00:30:00] you're asking it to do something like line editing or copy editing, that's a different process. It needs to be very clinical in going word by word, letter by letter and focusing on each individual letter to make sure that it's catching all of those tiny errors.

And right now it's not able to give it's not able to do that job as well when you give it a large amount of text than when you give it a smaller amount of text, just because of the way that this, the technology operates. So it depends on the task. Some things, yeah. You can give it the full manuscript.

Other tasks you want to "chunk it" is what we call it in the biz. [00:31:00] You want to break it into smaller pieces and give it those, and that's what's going to make it more accurate. But I do see that people who are have tried AI for like copy editing and are unimpressed, it's often because they've done something like that or they've given it t oo much text to work with at one time, or they are not being specific enough in their prompts. They're saying something like, edit this text. And we know they know what they mean, but, the AI doesn't. That could mean so many different things to it. It could mean grammar, spelling, punctuation. It could make much wider changes and broader, deeper changes than they, they were thinking. But if you tell it very [00:32:00] specifically to change only grammar, spelling, punctuation, then you're much more likely to get the types of changes you want than if you use a broad and vague prompt.

 Steph Pajonas: I tell people all the time, the AI cannot read your mind. It cannot, yet, at least. Let's hope that we're saved from that for a little bit longer, right? The AI cannot read your mind. So if you just say edit this piece of text it's not gonna really, doesn't really know what to do, right? So you have to say things like, edit this piece of text for context and clarity, c heck syntax errors, those sorts of things. You have to actually be verbose in your prompt and detailed so you get back something that is actually useful, right? This happens to, to a lot of people, especially their first time using it, which is why they come to people like you to teach them how to actually edit with these tools.

It's not just dump a manuscript in and say, [00:33:00] fix it. You give them tactics and w ays of prompting to get the kind of results that they want. And you've been working on that for some time now, so I'm sure that that they're really excited to actually get some good guidance on how to edit using these tools.

 Danica Favorite: Yeah, I think it's really cool, like what you were saying, Steph, about the tactics, because you know what you said Erin, a bout, for this task, it can do the whole manuscript, for this task, it can't, and really understanding that different things, different tasks require different tactics.

They require different models, they require different prompting. And it isn't a one size fits all. I think that's really good for people to keep in mind. Everything is situationally dependent and understanding what those situations are. Do you have any other really great tips or things that you want people to really examine and look for when they're editing their manuscripts using AI?

Any [00:34:00] other really great tips? I know that transparency is a big concern with editors. They don't like the thought that all of these changes are happening in this black box of AI and they can't see it and it's understandable. At the end of the day, it's the human's name who's on the project. We're the ones who are listed as the editor, if the author is so kind as to name us in the acknowledgements, so we want to make sure that we know the changes that it made. We don't want to insert any mistakes, and there are some ways to get that transparency. It can be as simple as you can ask for a downloadable dot docx file a word file and some models and tools can create these.

 Erin Servais: If not, you can copy and [00:35:00] paste the output into a Word document and then do the compare documents function to get track changes. You can see what happened. There's a browser plugin for ChatGPT, that's called editGPT. And that shows you directly in ChatGPT, the changes that it made. So there's that too.

You can ask the AI to list the changes that it made. It's not always a hundred percent whenever it lists it, so know that. But it is one way to do it. And, where I explained in line editing where you're having it do one type of change at a time, show its output. I like to say and list the changes. I even say put the changes in a table where you have the change and then the reason why you made the change in a nice, easy to view format.

So [00:36:00] each time you can read the output, look at the table, make sure that you agree with everything before you move on. Which is why for line editing, you're not gonna want to put in 50 pages 'cause that would be a time consuming task, and we're looking for AI to help us be more efficient and effective in our work.

So using it for smaller pieces with line editing helps there. But those are some ways that you can get that transparency and feel more secure knowing the changes that it made.

 Danica Favorite: I really love that. I love that idea of asking it to put in a table. I'm totally gonna start using that suggestion now because yeah, like sometimes you, you forget what is going on and I think that's really important to have that quick glance.

I'm like, yeah, that, that's going in my toolbox. So our last question, I know we've talked a lot about  Notebook LM, but do you have a favorite AI tool? Is it  Notebook LM? Are there other tools that you really [00:37:00] like?

 Erin Servais: There are, they change. So Notebook right now is one of my favorites, if you couldn't tell. I also have been really excited mildly obsessed with making custom GPTs, which are mini versions of ChatGPT that do a single task, typically. You can do something similar in Google Gemini and in Copilot as well. I have some custom GPTs that I've built for editing and I like, you don't have to, but I like to give mine big personalities. It makes them more fun to work with. So you can give them a role to play a persona.

So my favorite one for editing, I've named Mr. Casper. And Mr. Casper is named in honor of a real Mr. Casper, who was [00:38:00] my speech and debate coach from high school. And he was pretty stern and a little gruff in his feedback. He wanted you to do well, but he was very, had a particular way of relaying feedback.

So I, I built a GPT that had his personality and it is a really ruthless line editor that will just tear apart your text and really does give you a version that is better. But it is interesting the way that he does it. And I've even had him work... In the instructions I said, you must find a way to work into the feedback every single time a reference to how you coached the 1999 national winner in the humor [00:39:00] category in the National Speech and debate c ontest. I can't remember exactly what it was called. And he does, and it's wonderful which is not related to editing, but showing that you can make these a little fun.

Custom GPTs are incredible. And one way that I think authors could find this helpful is by creating an ideal reader GPT. Maybe you've already come up with a profile for your ideal reader. I know a lot of authors like to do that, where you have their name, where they're from, what their favorite pastimes are, what kind of books they like to read.

You could easily turn that into a GPT. Basically, n ot much more than copying and pasting what would be in your document, your ideal reader document already. And then you can have this chat bot that is that reader and you can give them a chapter and ask [00:40:00] what they think of it, ask for their feedback, and they can be that alpha reader for you, and help you refine it before you give it to your editor or give it to whoever your first readers are. And maybe be able to have something that is more improved than it would've been otherwise before when you got started. 'Cause isn't that wonderful? If you can use it to improve your story before you hand it off to the human to do their work.

 Danica Favorite: Yeah. I love that. I think that is awesome. I love that idea of having your chat bot be your, your ideal reader, and that way, if you're hitting the mark, because isn't that what we all wanna do is hit the mark with the readers? So I don't know about you Steph, but I am sitting here mind blown with all of these amazing tips.

I think I actually was slyly looking up your editing class and thinking all right. I may just have to take Erin. I, no, I may, I'm gonna sign up for your [00:41:00] next editing class, because I think that, just the things that I've learned simply about editing. I think, as an author, I know how to read, I know how to write, but some of the editing, since that is my least favorite thing to do, I'm like, okay, maybe I should educate myself on this a little bit more.

So thank you. Thank you, Erin. Really happy we met. Happy to have had that moment where you came up to our table and were so excited to meet me and I'm like. You don't know this part, but like as soon as you left the table, I'm like texting Steph. Guess what? Guess what? Because those moments are really great.

And for those of you who are listening ...please, if you do see us at a conference or somewhere like that. Really do reach out and make sure you tell us because it is very exciting for us. And I promise the second you turn your back and walk away, we are gonna be texting each other going, oh my gosh, isn't this so cool?

Because this is why we're here and what we love to do. We really do love knowing that we're helping people. So thank you. Thank you, Erin.

 Erin Servais: Thank you so much for having [00:42:00] me.

 Steph Pajonas: Thank you.

Do you, before we wrap up here, you should give us some URLs and places to send people so that they can find your courses and all of your information online.

 Erin Servais: Yes, I'm very happy to share that. You can learn more about my AI for Editors courses at aiforeditors.com. And if you are with an organization who may be interested in custom training, you can find information about that at servaissolutions.com.

Let me go ahead and spell that last name for you. It's S E R V, as in Victor, A I S solutions.com.

 Steph Pajonas: Perfect.

Perfect.

Thank you for coming, Erin. We're so grateful you came and talked to us about using AI for editing, which is a topic that so many authors wanna know about.

So I'm glad that we're able to bring it to them. So everybody who's been [00:43:00] listening drop by bravenewbookshelf.com. Read the show notes. We'll have all the information in there, including any URLs that we mentioned in today's course. And subscribe to us on all the usual podcast apps. Maybe go on over to Apple Podcasts and leave us a review.

I, I still think we don't have any, it's been over a year. I know you guys are out there listening and love 'em, because the stats say that you are. So come leave us a review. Anything else you wanna tell them before we leave Danica?

 Danica Favorite: The one thing I do wanna say is Apple Podcasts. We have had messages from listeners who were having a hard time because it wasn't showing up right on Apple Podcasts, and I know Steph has been working diligently behind the scenes to fix it.

So it should be fixed. So if you abandoned listening to it on Apple Podcasts please go back and try again. But also remember that for those of you who are listening, [00:44:00] if you ever want to see our lovely faces or see Steph's beautiful beach house, I've been admiring it the whole time with great desire and longing because it's so beautiful.

Please make sure you are watching us and liking and subscribing on YouTube as well.

 Steph Pajonas: Definitely we, we love putting our wonderful faces on YouTube for you guys. All right, everybody, thank you so much for listening. We'll be back next week with even more guests and cool tips and all that good stuff.

So from everybody here, we'll see you later. Bye

 Danica Favorite: Bye

Thanks for joining us on The Brave New Bookshelf. Be sure to like and subscribe to us on YouTube and your favorite podcast app. You can also visit us at bravenewbookshelf.com Sign up for our newsletter and get all the show notes.