Skip to content

Episode 28 – The Role of Community in AI-Driven Authorship

Listen

Elsewhere Online

Apple Podcasts | Spotify | Amazon Music | YouTube | iHeartRadio | RSS Feed

This episode focuses on critical discussions about using AI responsibly, overcoming fear of backlash, and leveraging AI tools effectively in your author business.

Understanding the Role of AI in Publishing

The episode kicks off with an important distinction between two types of AI:

  1. Machine Learning Algorithms: Used in backend operations like data handling or speeding up website performance.
  2. Generative AI: Tools that create text, images, music, or other forms of media based on user input.

While some may be wary of generative AI’s role in publishing (such as assisting with writing or editing), both hosts stress that this technology is already embedded in numerous industries — from insurance claims processing to hospital systems. The key takeaway? Instead of fearing it, authors should educate themselves on how to use it responsibly within their ethical boundaries.

Current Events: The Backlash Against Authors Using AI

One topic that sparked passionate discussion was a recent incident involving an author who unintentionally published a book containing unedited generative AI text. While the mistake led to backlash from anti-AI groups — complete with one-star reviews — the hosts emphasized an important lesson: a few vocal critics cannot “cancel” your career if you continue producing quality work for your readers.

Mistakes happen (even without using AI). Whether it’s a typo or an overlooked edit note, these errors are universal across traditional and indie publishing alike. The real takeaway? Learn from these experiences and move forward confidently.

The only person who can end your writing career is you. Stay focused on delivering great stories to your audience. How you get there is entirely up to you.

Building a Supportive Community

For authors feeling hesitant about openly using AI due to potential criticism, Steph and Danica recommend joining supportive spaces like their AI Writing for Authors Facebook group. Here, you’ll find encouragement from like-minded individuals who embrace innovation while respecting each other’s unique creative processes.

Danica shared her personal journey of finding her “tribe” after facing challenges in her local community during a difficult year. Her message was clear: Your voice matters, and there’s a supportive network out there ready to help you succeed — AI-positive or not.

Practical Applications: Using Transcriptions and Prompts Effectively

One major highlight of this episode was showcasing how both hosts utilize generative AI tools in their work lives. Below are examples shared by Steph and Danica:

1. Steph’s Experience Creating Documentation with Gemini

Steph recently worked on documentation for FFA’s new tool called PlotDrive — a cloud-based platform designed to help writers brainstorm and write story ideas more efficiently. Here’s her process:

  • She used transcripts from Zoom meetings and YouTube videos demonstrating PlotDrive.
  • Leveraging Google’s Gemini LLM (Large Language Model), she developed step-by-step documentation over four days.
  • Gemini helped generate outlines, refine sections into readable guides, suggest screenshots for clarity, and ensure accuracy through iterative edits.

Transcriptions are invaluable for repurposing content into blogs, newsletters, Patreon posts, or even full-fledged product documentation!

2. Danica’s Dictation Workflow & Troubleshooting Challenges

Danica often uses dictation apps while driving long distances for work or personal errands. Her typical process involves:

  • Recording thoughts via her phone’s voice memo feature.
  • Uploading transcriptions into ChatGPT for cleanup (removing speaker references, correcting punctuation).

However, she encountered issues this week when ChatGPT failed repeatedly due to incomplete responses or misinterpretations, forcing her to manually edit transcripts instead.

Lessons Learned:

  • Always have backup tools ready (e.g., Microsoft Word’s transcription feature).
  • Accept that sometimes you’ll need hands-on involvement (“Don’t be a T-Rex,” as Elizabeth Ann West from FFA puts it).
  • Community input can provide fresh prompt ideas when workflows stall—another reason why having a support network matters!

Encouraging Open-Mindedness About AI

The hosts reiterated that there’s no “one right way” to write or publish books, and this extends to whether you choose to use AI tools or not. Respecting diverse creative processes enriches the industry as a whole.

They also highlighted how advancements in accessibility, like automatic captions powered by generative AI, are improving inclusivity for readers with disabilities. These benefits underscore why staying open-minded about technological progress can be so impactful.

Upcoming Events & Opportunities

Both Steph and Danica are actively involved in conferences over the next few months where they’ll discuss topics like metadata generation and book cover design using PublishDrive’s tools:

  • AI Writing Summit
  • Women in Publishing Summit
  • Future Publishing Mastermind (New Orleans)

Additionally:

  • Participate in PublishDrive’s ongoing AI & Publishing Survey, which closes soon.
  • Follow Brave New Bookshelf on social media platforms like Facebook and YouTube for updates!

Links Mentioned:

  1. PlotDrive
  2. PublishDrive
  3. AI Writing Summit
  4. Women in Publishing Summit
  5. Future Publishing Mastermind
  6. AI Writing for Authors Facebook Group

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 a episode of Brave New Bookshelf. I’m Steph Pajonas, CTO of the Future Fiction Academy, where we teach authors how to use AI in any part of their business. This is going to be a little duo show today, just me and my lovely co host Danica. Danica, how are you doing today?

Danica Favorite: I’m doing great. Thank you. I’m Danica Favorite. I am the community manager at Publish Drive where we have a complete ecosystem to help authors on every stage of their publishing journey from helping them craft the perfect metadata and book description as well as our new product AI generated [00:01:00] covers then to distribute your books and also then split royalties once your books are selling.

So a lot of cool tools. And I love that we have a combo of AI and non AI tools and regardless of which part of our product you want to use, there’s something for everyone. Use one, use them all. There’s a place for you.

Steph Pajonas: Yes. Same, same at Future Fiction Academy. We’ve got education and we’ve also got tools for writing.

We’ve got Raptor Write and then we just launched like last week. I’m not sure when this podcast will go out. Probably very soon. Plot Drive. So that’s at plotdrive. com. And that is a dual business that we have with the guys from Realm Chef. So we partnered up with them and we came up with this tool that’s sort of like Raptor Write on steroids.

It is a, it’s a big tool. It saves to the cloud. It has a lot of cool features, including mobile friendly, et cetera. So there’s a lot there. I mean, between FFA [00:02:00] and PublishDrive, we certainly offer everything from soup to nuts, right?

Danica Favorite: Absolutely, and I think that’s great because, there are so many different things out there, and sometimes it’s hard to know what to do or how to do it, and I think between the two of us, we really work to give something for everyone, no matter where they’re at in the publishing journey.

Steph Pajonas: So we’re decided that we’d come together to do a little duo show today. Mainly because we see a lot of stuff going on in the social media verse. People talking about, about writing with AI as usual and also publishing with AI. We’ve seen a few things happen in the past few weeks that have, made me feel bad.

Made me feel bad for a few people. And also get me fired up about what’s been going on. Do you want to brief on the situation, Danica, or?

Danica Favorite: Yeah. So I’m going to talk about briefly another situation, but I think that kind of goes into some of what we’re going to talk about.

So.

Steph Pajonas: Go for it.

Danica Favorite: One of the [00:03:00] things that I can’t remember if we’ve mentioned this or not. We did. We mentioned this on our last episode, but the AI and Publishing Survey that Publish Drive is doing. It is still open. We’re going to keep it open through the end of the month, so when this episode launches, you’ll have like another day or two to fill it out.

I really encourage you to fill out this survey. I know it is long. The feedback I’ve gotten is it’s way too long. I understand that. Initially, we’re hoping to get like demographic information so we could see like where people were on the publishing journey as, as compared to what their thoughts on AI were and see if there was any correlation.

We’re trying to see what kind of demographic correlations there were, but it made for a really long survey. So when we do this again next year, I promise it will be shorter. We’ll ask fewer demographic questions just so that way we can make it faster and easier for people to answer. But I’ve gone through some of the answers, because I couldn’t help but peek, and they are all anonymous, so I don’t know who said this, but there are a lot of [00:04:00] people who still are giving really ignorant responses to things about AI.

And As much as we think we know all there is to know about AI, I just really encourage people to continually educate themselves because it was really interesting that some of the responses that I got were just hilarious. Like, one of the people said, you are clearly an AI positive company and I refuse to do business with any company that uses AI in any form.

And I just wanted to laugh about that because I’m like, okay, to this person, I just want to say, awesome, but you’re not going to do business with any company. Because every company, whether they admit it or not, is using some form of AI. And maybe they don’t fully realize it, maybe they don’t understand that what they’re using is AI, but it’s being used in everything.

Like in the last episode, I was talking about how my mom was so frustrated about the AI on her cell phone and wanted to have it taken [00:05:00] off. Well, you can’t. The genie is out of the bottle as has been said by multiple people, multiple places, the only real answer is to learn how to use it appropriately in your circumstances.

And so, like, respectfully person who refuses to do business with any company that uses AI. You’re not going to do business with anybody and I’m really sorry.

Steph Pajonas: Good luck. Good luck. Right? There’s a segment of people who want to distinguish between AI as in machine learning algorithms, greasing out the back ends of your website so that they run faster or whatever it may be, handling data, etc. And generative AI, which is generating text, generating images, generate music, generating sounds, whatever it may be. Now, there is somewhat of a difference between those two, but you would be surprised how many companies are using generative AI on the back end, and they are things like your insurance company.

Your insurance company is using [00:06:00] generative AI to look at claims and summarize them and send them on to claim adjusters and whatnot. It’s quite possible that your medical company, hospitals are using it as well for these sorts of things. It is being used in ways that you are unaware of and they are not required to disclose that to you.

So we’re now at this point where it’s baked into a lot of businesses and it’s getting into all of your operating systems on your phone, and there are plenty of ways that you can turn a lot of that off, eventually it’s going to be default. There’s really going to be no way around it.

So it’s really important for us to determine how we’re going to interface with this technology so we can use it responsibly and use it within our own ethical boundaries.

Danica Favorite: Right. And I think also understanding that and realizing, you know, people are in uproar about generative AI being used for writing, for example, because obviously that’s the focus of this podcast. [00:07:00] But honestly, because I work with generative AI, I know the limitations of generative AI. I am a little concerned about it being used on my medical information, but it’s there. And at least when I’m aware of it and I know what’s going on and I know how generative AI works, when I get something weird from my insurance company or from the doctor, when I go and speak to a live person, I know exactly what’s going on and I can clarify that and I can realize that It wasn’t a live person behind that decision.

And so I can work with them to resolve the issue better.

Steph Pajonas: Yeah.

Danica Favorite: As opposed to like freaking out because this weird thing came up and you don’t know how or why, and then immediately you can go, oh yeah, this is generative ai. Okay, I see the problem. Boom. I can have a conversation. I think at least the more we understand and know we can start to interpret some of those things.

Steph Pajonas: I agree. I mean, I was just on a support [00:08:00] chat the other day. And in fact, I’m sitting here thinking, which company was it? Because I can’t even remember. I definitely have to get on support chats. It’s like almost a few times a week at this point. And it just kept repeating the same thing over and over again.

And I was like, your AI is broken. I can tell. I can tell. It’s repeating the same things over and over again until it could send me off to a real human being.

This is all about education and understanding what we’re up against and understanding how to deal with it as we move forward. Because we’re in a time of transition. It’s going to be a while before this is an everyday, easy thing to slot into our lives, but that time is not now. It is sometime in the future, and in the meantime, we have to learn to deal with it and educate ourselves about how it is being used around us and with our own information and our own data.

Danica Favorite: Absolutely. So, I think that’s a great segue into the situation that I think you’re talking about. So why don’t you come and give [00:09:00] that summary? Because I think it is really important, and it’s a valuable conversation. And I think it is something that listeners will really appreciate and understand some of the nuances.

Steph Pajonas: Okay, so a situation happened I’m not gonna mention the author’s name because I don’t think that this, this poor woman needs any more people ganging up on her, okay? She published a book and it turned out that in her revision part of her process.

She was using I believe maybe GPT to help her revise some sentences and she accidentally left the full prompt response, which started with, certainly I can help you revise that sentence. And here it is. And it did a great job. Which is kind of funny. I read the sentence and the response from the AI because they were both in the book.

What’s interesting, what is interesting is this particular book had gone through edits, had gone through a beta team, had gone through an ARC review process, and had ended up in the store [00:10:00] and in the top 100 of all of KDP before somebody noticed. And I always feel really bad for authors who end up with typos in their books or they’ve put the wrong character name in there.

God, goodness knows I’ve renamed my characters and accidentally, like,

Danica Favorite: oh yeah, like, you know that meme, my favorite meme, is like the one celebrating the typo, that went through all of those rounds of edits and still made it into the final copy because every author can relate to that and these are with human copy editors.

These are with all kinds of revisions and rounds of revisions. And sometimes, you know, there’s just that little typo that could, and we just have to celebrate it sometimes, you know.

Steph Pajonas: You do. I remember if I want to change the name of my character from Robert to Marcus, I always have to remember that there’s also these ones floating around out there, the Robert apostrophe S, and I have to search for those too to replace them.

Danica Favorite: And we were joking in the AI [00:11:00] for Authors group about this, gosh, probably within the last week as well, where you change the character’s name and you find and replace, not realizing that, you know, Chris is actually part of Christmas and Christian and whatever. And so you change the character, Chris, and suddenly you’ve got Robert ness, and like sometimes it just happens, right? Editing mistakes are there. And yes, I’ve seen these in traditionally published books. I’ve seen these in all aspects of published work. I think it has nothing to do with the professional quality of a book.

It just happens sometimes.

Steph Pajonas: It does, it does. But the big news here was that it was obvious that it was an AI response, right? It was obvious that it was an AI response that was still in there and hadn’t been deleted, and this really angered the anti AI community.

It really did. They went and they one starred this woman’s book [00:12:00] and they raked her over the coals in almost every single group that they could find. She felt bad. She pulled it down. Maybe she was reading through her book the last minute thinking, oh, does the scene need a little bit of help?

And she did a little revision really quickly and forgot to erase the prompt response.

So, I feel for people in these situations because I have been there, you’re on a deadline, you’re trying to get something done really, really fast, really efficiently and still produce a quality work and you just, you screw something up.

I have done it many, many, many, many, many, many times. But I think that the thing to remember here is that even though the anti AI community completely went after this woman and left one star reviews on her books and posted about her in a bunch of Facebook groups, that book that she took down still did very, very well in the store.

It got a [00:13:00] lot of great reviews before anybody came to One Star bomb it, and the rest of her books are doing really well still. So this is a lesson to all of the people who worry about being cancelled for using AI. No one person or even a little group of people who are vindictive and bullies can cancel you and your career.

Let me say that again. They cannot cancel you and your career. All right? You are producing work that your readers are interested in, they want, they love, and you’re going to give it to them however that works for you and your process, whether you’re doing some AI generation and you’re editing or you’re using AI to help you edit prose that you wrote yourself, whatever it may be.

Your readers want what you can give them, and a few people who are very loud, very bullyish, are not [00:14:00] going to end you. Okay?

Danica Favorite: Yes, and I think that’s important to note, and I want to actually take us back a little bit in history on this, because I’m not saying this is right. However, some of you may remember some of us are old enough to remember, some of you are not, and this is why I want to bring this up, is, years ago.

I think it was like 20 or 30 years ago now. I was going to look up the exact date and then I forgot. So anyway, Nora Roberts was plagiarized by Janet Daly. This was proven in court. Nora won the court case. Very important to note. Janet Daly continued to publish books and continued to have a successful career.

I do not in any way condone what Janet Daly did. And I’m one of the people who said, Ooh, she plagiarized Nora Roberts. I’m never buying her books again. I was one of those who said that, but you know what? It didn’t kill Janet [00:15:00] Daly’s career, like she lost the lawsuit about plagiarism, proven plagiarist.

Steph Pajonas: I remember that .

Danica Favorite: Didn’t hurt her career. And again, I am not saying this to condone plagiarism because I really, really do not like plagiarism. But no one is going to be able to cancel your writing career just because you use AI, which is way less intrusive than plagiarism.

And to be very clear to all of those out there, using generative AI is not plagiarism. If you have questions about that, we’ve discussed this on multiple episodes of how generative AI works, why it is not plagiarizing.

Using generative AI in and of itself. Not plagiarism. Okay. Off the soapbox. Sorry.

Steph Pajonas: I hear you. I hear you. And you know what?

Let’s, we need to acknowledge the fact that Janet Dailey, the only person who could have ended her career was her. She could have stopped, [00:16:00] right? At that point, she could have lost that case to Nora Roberts and said, I’m done. I don’t want to do this anymore. And then she could have ended her career, but she chose to go on, obviously, and still have a career after something that big of a profile of an event.

So, like, the only person in charge of your career is you, plain and simple.

Danica Favorite: And I will say, if you are someone who is fearful of what that looks like to use generative AI for your writing career or any part of AI, I don’t care if you use generative AI or not, whatever you’re using AI for in your career, if you’re afraid that people are going to cancel you or you need a safe space, we have our AI Writing for Authors group on Facebook.

Please join us. We will be your safe space, you know, in the very beginning, I can remember Steph and I, we talked about this where people were getting attacked for using AI and a lot of us were just going out and buying their [00:17:00] books just to support them. Now at this point, we can’t like there’s just too many, but you know, like seriously, there’s a whole community of authors who will help you.

And who will support you and will stand behind you. And, you know, that’s one of the things I want to just really remind you of, and this is kind of on a personal note to go off a little bit of a sidetrack, but, you know, last year was a tough year for me. I’ve said that a lot. I can finally just say it was hard because I got divorced and I lost so much of my local community because of how nasty the divorce got.

And that’s why I didn’t talk about it because I didn’t want it to continue to get nasty. But what came out of those ashes was finding my true community and finding who my real friends were, and finding people who are really there for me and who really care about me. And that is what I see in the writing community as well, is that, when I see these anti AI posts, when I see people who aren’t safe people to be out about AI, I know [00:18:00] those aren’t my tribe.

And I know, like, Steph is my tribe. You know, Steph and I would have never become friends had it not been for AI. This is how we met and how we bonded. And she writes totally different stuff from me. She writes genres that I wouldn’t ordinarily read and now I’ve gotten to know her and become her friend.

I’ve read her books. She’s a delightful writer and I’m really grateful for that. I’m grateful that having to move into another space to find my actual tribe has really made me find people that are really there for me. So, you know, those of you who are afraid, don’t be afraid.

Your people are out there. We’re here to help you and we’re here to guide you because what you have to write, what you have to say, the stories that you’re putting out, however you need to put them out, they are important. They matter and you matter to us.

Steph Pajonas: I agree. I found my tribe [00:19:00] definitely with the AI positive people.

And I really appreciate all of the friends of mine who remained AI just neutral about things and, and just told me, you do what you need to do, Steph, and I’ll be here. And I really appreciate that too. You know, there is a place you can stand where you don’t have to do either or if you don’t want to. I’m really grateful for the fact that we’ve built up a good community of people who are doing this and are remaining open minded and excited about all the things that are coming.

Danica Favorite: I think you really do because there is a lot of confusing stuff out there. I think what has served all of us in this community the best is that open mind because the newest and latest thing will come out and automatically you jump on the bandwagon And what we’ve learned to do, because there’s so many new great things coming out, is we’ve really learned to be discerning in what makes a good AI, what we’re [00:20:00] looking for in AI, what is valuable, all of those things.

And it’s okay to not have an opinion. It’s just not okay to criticize somebody for not having the same opinion as you about AI. If you don’t want to use AI, I totally support that, because I support you. I think that that’s your right to choose, and that’s wonderful. If you feel like that’s important to you, great.

But we just have to stop being negative to people who aren’t doing it our way.

Steph Pajonas: There’s no one right way to do anything.

Danica Favorite: You know, there really isn’t. Like, I mean, I guess there’s probably a one right way to diffuse a bomb.

Steph Pajonas: Yeah, maybe.

Danica Favorite: Yeah, you know, but like, in general, in life, like, I think the older I get, there’s this wisdom of realizing that everyone has their own process.

And you have to respect that, and you have to respect who they are and [00:21:00] what they do because it is different for everybody. And particularly, I know with AI, we have a large neurodivergent population who use AI, and I’ve become friends with more and more neurodivergent people. To realize that, there’s nothing wrong with them.

There’s nothing wrong with us. , I’ve been told my whole life, Oh, this is wrong with you. No, I just have a different way of thinking. I have a different way of doing things. And it’s not that it’s wrong. It’s just not neurotypical.

And so let’s take that learning into everything else we do. And apply that because I think that that for me is the real beauty of AI.

Steph Pajonas: I agree. All right. So we’re not going to dwell too much on this topic. We have covered it before. We just felt we had to cover it again.

Danica Favorite: current events just kind of made it pop up.

Steph Pajonas: Yeah.

Danica Favorite: So every once in a while, it will pop back up. But, you know, we just [00:22:00] want to be out there encouraging you. And so one of the things we’re going to do today, actually, you know, we’re telling you guys constantly, hey, let us know if there’s something you want us to talk about on the podcast.

So I have this email from Matty Dalrymple and she works with Alli and helps with some of their AI stuff. She says thank you Danica. I wanted to reiterate how much I’m enjoying the Brave New Bookshelf podcast, and I’m trying to encourage authors to be at least AI curious in my work as a campaigns manager for the Alliance of Independent Authors, which I thought was really great.

So thank you, Matty. We really appreciate you. We appreciate that you took the time to write me. And it’s kind of a long email, so I’m not going to read the whole email, but she says, I think it would be super useful and fun to hear specific examples of the prompts people use and how they refine those prompts to get the output they need.

Ideally using an AI tool people can experiment with for free. It could be fun to have a periodic spot about my last five prompts where you and Steph could talk through the very specific ways you’ve used and [00:23:00] optimized AI.

So Matty. We loved this idea, and that is what we’re talking about today.

That is going to be where Steph and I go with this. We don’t know that we’re going to get to five prompts because we did go a little bit long. And also, I think the explanation of some of the prompt stuff we do is also going to take a while. a little bit longer so we won’t get to five.

Anyone else, we’re here, we’re open. Bravenewbookshelf. com is our website and you can contact us through there or you can contact us on Facebook, Brave New Bookshelf. Or leave a comment on our YouTube videos. So, yeah, lots of places for you to leave comments and feedback. We would love to hear it. So, Steph, let’s start with you and your prompts because I know you’re doing really cool stuff.

Steph Pajonas: Yeah, I think that basically this might be a, hey, what did you work on this week using prompts? type of segment. So I was just speaking earlier about the fact that we’ve got a new piece of software out called Plot Drive. Plot Drive it’s, it’s a little bit different from all the other writing tools.

We give [00:24:00] people the ability to really freestyle and make buttons that run prompts and we had our team of developers working on right up to the very, last moment that it launched. They were working on it until 15 minutes before launch when they were like, okay, it’s ready to go.

So as you can imagine we open up the tool and we start showing it off on YouTube. And my business partner, Elizabeth was on there showing it off, making prompts. getting it to write for her, do all this kind of stuff. And then people said, great, where’s the documentation? I want to take a look at it so that I can understand how to work this.

But, the documentation was not available because of the fact that we had been working on this software right up until the very launch of it. Documentation usually falls in my lap at the FFA. I do a lot of documentation for people. I really don’t mind. I think it’s a fun thing to do even if it is kind of stressful [00:25:00] trying to get it done under a deadline.

I love to get screenshots and put circles and arrows on them and point to things and say this, this, do this, in this step and click on that button or whatever it may be.

So I have a way that I go about making documentation for things, especially at the FFA. And it involves getting somebody else to show it off.

Especially on YouTube or in a Zoom class that we record and we get a transcript. once I get transcripts to things, I am all set and ready to go. I choose a LLM that has a large context window, and in this case, I chose Gemini. Gemini 2. 0 Experimental, which just happens to be mostly free.

I believe you can use it through the website free for now because it is in an experimental phase. What I do is I give it all the transcripts of all the videos that we did on Plot Drive and then I started asking for help on [00:26:00] it. So I’m going to go back and I’m going to read my first prompt to you guys that had the transcripts attached to the conversation and I said,

Hello, I’m working on documentation for a new AI writing software we’ve developed called Plotdrive. It’s brand new and we have had a few video sessions on YouTube and Zoom to show people how to use it. But now it’s time for me to make step by step written documentation for the tool for everyone. I need to show people how to use PlotDrive from the start of a fresh account. To prompting their very first text, I’m attaching the transcripts of the videos to this conversation and a chat transcript from a zoom meeting, and I want you to help me build the documentation for it. Let’s get started.

And at that point, I’m able to get usually an outline. of the documentation with ideas for each one of the sections from GemIIni and then I start drilling down with Gemini. I said let’s work on section one, Getting Started. We’re going to go through each one of these suggestions that you made and we’re going to flesh it out [00:27:00] into like blog post type documentation where you’re going to give me headers and sections of text and you can make lists as well.

Give me call outs to where you think I should put screenshots that would be really helpful. And then I just start working with Gemini back and forth. It gives me stuff. I make some edits to it. I make the documentation. We move on to the next one. If it makes wrong inferences based on what it had read in the transcripts.

I correct it. I say, no, you can’t do that with that kind of button, but you can do it with this kind of button instead. It corrects itself. It ends up in the memory so that it doesn’t end up making that mistake again, which is great. And I keep doing that until the documentation is done. So I have produced about 15, 16 pieces of documentation for PlotDrive.

I think it might be even more, closer to 20. And it took me about four days to get it done. That’s not [00:28:00] four straight days. That’s like I work on a like an hour and a half here, an hour and a half there, sometimes in the evenings when I had a little bit of time because I’m basically running five businesses right now.

I only have so much time to devote to these kinds of tasks. So that’s how I used an LLM this week doing something really cool. I often suggest to people that if you have to show something to somebody, if you’re more comfortable, showing it in a Zoom where you’re just talking it through or maybe you’re out on a walk and you can just talk it out by yourself and talk into your phone.

Do these sorts of things because once you have that, you can always take the transcript and then get AI to help you make it into something a little bit better for your product or whatever you may be doing.

Danica Favorite: Okay, so I have two questions on that for the transcription, what did you guys use for transcription?

Steph Pajonas: Okay, so Zoom does a lot of the transcribing on its own within [00:29:00] the account. And also when you upload a video to YouTube, it will do a transcription because it wants to make subtitles.

And it will also do it on lives too. So for this particular task, I had two YouTube lives that we did showing off the application. And then I think I had a zoom meeting as well that we had. It was recorded to the cloud on zoom and it made a transcription of it there. You can also, if you don’t have access to those kinds of tools Descript can help you make a transcription of your video.

If you use your phone Voice Memos on iOS has automatic transcription built in. There’s a lot of things that can help you with transcribing.

Danica Favorite: Awesome. Awesome. So then the next thing is, how did you access Gemini?

Steph Pajonas: I accessed Gemini through TypingMind. It’s TypingMind. com for anybody who wants to go check it out. But there is AI studio from Google and I believe it is. Aistudio. [00:30:00] google. com. You can go there and you log in with your Google account and you can play with their tools there. There are some that are free or you get like a free couple of responses, but it’s definitely a good place to go and try it out.

Danica Favorite: Awesome. Yeah. I was just curious, cause I, I know you use TypingMind a lot and I wasn’t sure if this was through TypingMind or through Google directly. So again, just important for people to know the different tools we’re using.

I was working on a lot of transcription this week. Like Steph, I’ve had a super, super busy week. Publish Drive is doing a bunch of conferences over the next few months. We’re part of the AI Writing Summit. We’re part of the Women in Publishing Summit. We’ve got the Future of Publishing Mastermind in New Orleans. And I have presentations to give for all of them.

And we have our metadata generator, which is what I’m highlighting in all three of those, in addition to our other services, but our metadata generator, as well as our cover generator, which [00:31:00] we have very little documentation in terms of, no one has taught anything about our cover generator yet.

And so. I had to come up with a plan for that, and I had to come up with also a plan for some other things that I’m doing for marketing and stuff for PublishDrive.

I’m swamped. Like, I just have zero bandwidth, and I’m thinking, How do I write all these presentations? With my dictation process, I drive an hour each way to the chiropractor. My girlfriend’s group, we meet once a month to play some games, and that’s about an hour drive to their house each way. So I was like, okay, I can get some work done while I’m driving. I’m sitting there dictating. And as I’ve told you all before, this is my dictation process.

I dictate into my phone. I get the automatic transcription. I pop it into Chat GPT to clean up. It works like a charm.

However, this week, I guess I broke ChatGPT because ChatGPT is not behaving. I even went to the AI for Authors group to say, Hey, like, what [00:32:00] prompts are you using that are working? Because my prompts, my prompt style just completely failed me.

And so here’s my prompt. I said, I have a transcription of a voice recording that needs to be cleaned up, taking out references to speakers, have punctuation and spelling corrected, leaving the text intact. How do I prompt to do this?

And for me, what I have personally found is that when I’m finding that my prompts are going astray, if I tell the AI what I want to do, if I say to the LLM, I said, this is the task I’m trying to accomplish.

Help me create a prompt to accomplish this. Nine times out of 10, I get, the best prompt, best prompt results I have ever gotten, like I could get otherwise. And so, ChatGPT answers, you can use the following prompts to ensure the transcription is cleaned up properly while maintaining the integrity of the text.

Prompt, please clean up the following transcription by removing references to speakers, correcting punctuation and spelling, [00:33:00] and ensuring the text flows naturally. Keep the wording as intact as possible without changing meaning or restructuring sentences significantly. Then I edited it slightly for that because it’s not going to ensure the text flows naturally because then it changes it too much.

Really knowing what I want as someone who dictates. was really important because then I could adjust that prompt and say, no, you know, just keep the text intact. Let me tell you something. I ran prompt after prompt after prompt after prompt, and I kept getting summaries. It kept cutting it off, and I was using, like, I wasn’t using a huge, long document here.

It was still cutting off midway through and wasn’t giving me the last half of my text. And I’m thinking, why is this going off the rails? And literally I would go back and I would question it. And it would say, Oh yeah, sorry, let me fix that. And then it would do the exact same thing. Sometimes you just have to accept the AI is having a bad day that day and give up the ghost. I actually did go into Word [00:34:00] because Microsoft word also has a great transcription feature. So, and that is built into Word. That was built into Word prior to them putting, I think, what was it, Bard that they stuck in there?

Steph Pajonas: Yeah, BARD. Mm hmm.

Danica Favorite: Yeah. Even prior to that, it had a great transcription tool. I tried that and it wasn’t great, but it was better than ChatGPT going off the rails. I went in and I kept trying all the different LLMs. I just think maybe it was having a bad night, but I could not get any of my prompts to work.

So what I want to tell you with that, and, and this is really important because I think sometimes people will go in and they’ll try AI they think, Oh, you know, it didn’t work for me. This has been my process ever since ChatGPT came out. Okay, prior to this, I would actually, you know, just record it in my recorder and pop it in, download the recording but same process.

The AI models change constantly and what they can do changes constantly. And so I did get a few ideas of new prompts and I’m going to try them. I had a couple other people in the group [00:35:00] share. Well, this is the prompt that’s working for me and I think that’s why having a community is really, really important because it can be really frustrating. I finally turned in this report that I needed to turn in at 11 o’clock at night. Because I was fighting with the AI to do the transcription. And most of it I just ended up doing manually because I had the manual transcript, but it was the cleaning up of the transcript that was going off the rails. And I just finally cleaned it up because , I don’t have time. I fuss with the AI long enough. It’s time to just do it myself.

Steph Pajonas: Sometimes that is exactly what you have to do.

Danica Favorite: And I think that’s also a really important skill when you’re working with the AI to say, you know what the AI can’t do this task.

I just have to do it myself. That’s why the FFA has Rexy and the whole dinosaur thing is Elizabeth Ann West said from the very beginning, don’t be a T Rex. Sometimes you have to put your hands on the keyboard and do it yourself. And this was a really good reminder to me that even though I have my dictation transcription [00:36:00] process down really well, sometimes I just have to go back to doing it old school.

And that’s okay because It still is a good exercise. I still did save myself a lot of time by doing the dictation because I got to work while I drove.

Steph Pajonas: Right. And those ideas didn’t just like leave your head, which is what happens to me. I’ll have a great idea. I’m like, Oh, this is a great idea. I should write it down. And then I don’t write it down. And then I forget it. And then it’s just gone. So at least you said those things out loud, they were recorded, and you can then use it later, even if it’s not something that can be easily processed by AI.

Danica Favorite: And that’s really the thing, too, is that for me, because I’m so used to dictation, and it’s gotten so much easier now, I have found that that is generally my process anytime.

If I have an idea, I’m like, Oh, boom, open up the voice recorder and I start recording. And I’ve done this with friends as well, where [00:37:00] we’re brainstorming ideas and we’re talking about things where we’re trying to figure out solutions to a problem or whatever. And then I’m like, Hey, do you mind if I record this? We can get a transcript of this and I can send it to you and this might help you with whatever we’re trying to plan and whatnot. It’s so useful to be able to do that. But sometimes the AI is having a bad day. And you just have to say, okay, you can’t do it today. I can do it myself.

Steph Pajonas: There’s only so much you can do.

Danica Favorite: And sometimes it goes down. Like I was just reading ChatGPT was down earlier for a little while today.

Steph Pajonas: It was. It was and then I was trying to finish up my documentation earlier and Gemini was down too.

So I was like, Okay, ChatGPT was down. Gemini was down. Let’s move over to Claude. And then I did and I was working so great.

Danica Favorite: And People are so afraid of AI taking over is it’s still important to have the skills to do the things without the AI because in those moments when the AI goes down or the AI isn’t working, [00:38:00] then you can go ahead and do it yourself . You just recognize that you have the skills. And again, that’s why I’m not afraid of AI taking my job as a writer because it’s always going to need someone experienced with words to take over and to do some of the things that AI just can’t do.

Steph Pajonas: Mm hmm. And that’s a great place to end this conversation for today.

I’m hoping that each time we can do some of these shows together, we can talk about the things that we’re using AI for. I think that people are interested in that because we talk about the fact that we use AI for business and for writing and editing and all these kinds of stuff, but we rarely detail what those things are.

So I was excited to talk about this because of all of the stuff we’ve done recently. And transcription, I feel, is a really good lift for people, especially if you can grab transcriptions from videos you’ve done. That’s an easy thing to take it and make it into more content for you and for your [00:39:00] audience.

Think about all the different ways that that could be done. You could be having a reader’s event where you’re talking about your books or things that you wrote about or different themes in your work or any one of these things. If you had a reader event, you Talking off the cuff to people about it. You could take that recording, get the transcription, turn it into blog posts. You could turn it into newsletter content. You could turn it into, stuff for your Patreon. Transcriptions are a really easy lift and a great thing to do and use AI for, so consider it.

Danica Favorite: Also, again, you know, like going back to, we we’ve talked about things for helping people with disabilities. Now having captions and things like that is so much easier because of AI. Those transcriptions and everything, we don’t necessarily think of them as vital, but for someone who needs that, it’s absolutely critical.

So I love that we get to do that and we have those options now that are [00:40:00] easy for all of us.

Steph Pajonas: Yeah, I agree.

Danica Favorite: At least when the AI behaves. Come on, man.

Steph Pajonas: It wasn’t your week this week. I’m sorry. All right, well, coming up in our next couple of episodes, we’ll be having some guests. We’ve got them on the calendar.

They’re ready to talk to us about all the fun stuff that they’re doing. Is there anything going on with you before we say goodbye?

Danica Favorite: Yeah, I’ve got a lot of cool things that I’m going to be teaching on in the next little while for PublishDrive.

So if you are interested in our metadata generator, our book cover generator and even just our general basic services I’ll be teaching at the AI Writing Summit. I’ll be teaching at the Women in Publishing Conference. Steph and I will both be at the Future Publishing Mastermind in New Orleans. Those are things that are coming up in the next three months. That be on the lookout for those things. We have the AI and publishing survey that we’ll be closing like a couple of days after this episode launches. So if you haven’t [00:41:00] taken it, please do so.

Steph Pajonas: We’ll put the link in the notes for you guys. Okay.

Danica Favorite: Yep. That would be really super helpful. Please go like our Facebook page, go like our YouTube page. All of those things. Make sure you’re really liking and following. Subscribe to the podcast on your favorite podcast channel. So you’re notified about upcoming episodes.

And I’m really excited because the people I’ve invited who have said that they’re coming. I’m really excited about that. Some of them are people I’ve been trying to get on the podcast for a while and schedules didn’t match up and now they’re coming. We have so many cool guests coming up.

So please make sure that you are subscribed to everything. So you get notifications and like I said, if you have ideas and things you want to see on future episodes, please let us know, because, hey, Matty, this one was for you.

Steph Pajonas: And we’ll be talking more about those in the future, definitely.

All right, so come by BraveNewBookshelf. com. Check out the show notes [00:42:00] from today, the full transcript, leave a comment. Let us know what you’d like to hear about on the podcast as well. Like and subscribe, all that good stuff. And we’ll be back next time with more guests and more insights on AI and publishing.

Okay? All right. So we’re gonna say goodbye, right? Bye! See you guys soon.

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.

Leave a Reply

Your email address will not be published. Required fields are marked *

Brave New Bookshelf