Episode 42 - Innovative AI Storytelling with Russell Nohelty from Author Stack

Listen

Elsewhere Online

This week on Brave New Bookshelf, we had an enlightening conversation with Russell Nohelty, a USA Today bestselling author known for his expansive body of work, including 16 nonfiction titles and over 50 books in the Cosmic Weave universe. Russell joined us to discuss his journey from AI skeptic to a proponent, as well as his innovative use of AI in both fiction and nonfiction writing.

Meet Russell Nohelty, Prolific Author and AI Enthusiast

Russell Nohelty is a dynamic force in the writing community, with a diverse portfolio that spans genres and formats. From writing novels and comics to editing anthologies, Russell has done it all. His Cosmic Weave universe alone comprises 38 books, with more on the way. Recently, Russell has embraced AI as a valuable tool in his writing process, inspired in part by a conversation with Steph Pajonas about the potential of AI to democratize the writing process for those with different abilities.

Embracing AI: A Shift in Perspective

Initially hesitant about AI, Russell experienced a shift in perspective after discussing its benefits for accessibility and productivity. He realized that the act of typing isn’t the only measure of authorship. Storytelling is about curating an experience for readers. This understanding opened up new possibilities for Russell, allowing him to explore AI as a co-creator in his workflow.

The Craft of Storytelling: Where AI Fits In

A key takeaway from Russell’s experience is the importance of understanding the craft of storytelling. AI can assist in generating content, but it requires a writer’s expertise to shape that content into a compelling narrative. Russell emphasizes that AI tools can get you 70% of the way there, but knowing the end product you want is important for success.

Fiction vs. Nonfiction: Different Approaches to AI

Russell’s approach to AI varies between fiction and nonfiction. For nonfiction, AI excels at organizing concepts and expanding on ideas, making it easier to transform social media posts into comprehensive articles. In fiction, however, AI requires more guidance to maintain continuity and character development. Despite its limitations, AI is a valuable brainstorming partner, helping Russell explore new ideas and story arcs.

The Power of Riffing: Collaborating with AI

One of Russell’s favorite aspects of working with AI is the ability to “riff” with it, using the tool as a brainstorming partner to explore different narrative possibilities. This collaborative process allows for creative breakthroughs that might not occur when working alone.

Favorite AI Tools

While Russell primarily uses ChatGPT for his writing projects, he also appreciates Claude for its unique capabilities, despite its limitations with longer texts. He highlights the importance of selecting the right tool for the task at hand, whether it’s generating prose or conducting research.

Key Takeaways

Storytelling is about curating an experience for readers, not just typing words on a page.
Understanding the craft of storytelling is essential for effectively using AI tools.
AI can assist in generating content, but a writer’s expertise is needed to shape it into a compelling narrative.
Different AI tools have unique strengths and should be used accordingly.
Riffing with AI can lead to creative breakthroughs and innovative storytelling.

Resources Mentioned

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 an episode of the Brave New Bookshelf. I’m one of your co-host, Steph Pajonas, CTO of the Future Fiction Academy, where we teach authors how to use AI in any part of their process. We’re doing great, and over at the Future Fiction Academy. I was just talking with Danica before we started about the fact that we’ve just launched our Accelerator, which is a way for authors to come into the FFA and learn the basics of AI. And then if they are so willing, they can go straight into our Mastermind, which has tons and tons and tons of content. That is the fire hose of content. The Accelerator is more like the sprinkler.[00:01:00]

So we’re doing lots of fun stuff over at the FFA. We’re making sure that authors have these skills, should they want them, or they will need them at some point. We’re going to go ahead and provide that as best we can. So it’s been very busy around here and I know that life has also been busy for my wonderful co-host, as always, Danica Favorite.

How’s it going?

Danica Favorite: It’s going, gosh. Yeah, it has been busy, but oh, all good stuff. All good stuff. Very excited. 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 and publishers at every stage of their journey from formatting the manuscript to finding the right keywords and metadata, using our AI tools to creating the perfect AI cover. To getting that book distributed to a wide variety of sources. And then finally, once that book is published, we can help you with splitting your royalties. So a lot of cool stuff, and I love that we get to partner with the FFA on [00:02:00] this because of course we can’t help you write the book, but they can. And then they can help you learn how to do all the other cool marketing stuff and all of those different things that you need to do once that book is out there being published.

And I’m very excited about the accelerator because I know a lot of people have been wanting to do this for a while and so this is super cool. I was just telling Steph I’ve been the FFA forever, but I think I’m doing the accelerator too, so love it. Super excited.

And I’m also really excited about today’s guest. Today we have Russell Nohelty, he’s got the Author Stack, and so some of you may be following along with him. Steph and I just had a great time with him at his conference in New Orleans. Really enjoying hearing some of the ways that he’s using AI and how AI is helping this very successful, multi published author who’s been doing this for years.

How he’s been able to go from maybe being a little questioning about AI [00:03:00] to now saying, oh my goodness, look at all these ways I get to use it. So without further ado, I wanna present to you Russell Nohelty.

Russell Nohelty: Thanks so much for having me. I am USA today bestselling author, Russell Nohelty. I wrote all of these books and I’m, my keyboard is currently being lifted by all of my nonfiction books. So there’s 16 down there or something like that, and 50 books back there and comics and anthologies.

I’ve just done just so many things that it’s exhausting to literally just sum it up.

Danica Favorite: I think that’s awesome though. Like I did not realize, I don’t know why, ’cause I’ve been on a million recordings with you and I realize the people who are listening can’t see this, but he is, in fact, there’s a nice long row of books behind him and I don’t know why as many times as you and I have been on calls and things together, I didn’t realize those were all your books, which is really quite silly of me.

But here we are.

Russell Nohelty: That’s actually only all of the books in the Cosmic Weave universe. [00:04:00] So that’s 38 books there. I have an entirely different shelf for the other miscellaneous hundred or so other projects that I’ve either edited or been part of or written or there’s just a lot.

And I’m about. 500,000 words into a new series that is in the Cosmic weave too. So it’s, that’s like my big overarching universe of stuff.

Danica Favorite: I think that’s awesome. And I love having someone like you who has done a ton of writing both fiction and nonfiction. So it’s always good, I think, to have some of the experienced authors come on and talk about what their experience with AI is because people need to realize that this isn’t just one segment of authors.

That this is somebody who’s very successful, who’s using AI in publishing and using AI in their process. And I’m gonna okay, in a previous episode we had Monica Leonelle on and she was talking about, how Steph [00:05:00] had really influenced you in terms of your AI process. And what I really like about how you have been in the AI sphere is, I remember you were hesitant for a while, and then now here we are, you’re a guest of someone who isn’t hesitant.

I would love to hear how you’re approaching AI in publishing and how that evolution came about.

Russell Nohelty: Sure. Yeah, Steph, I was pretty negative about AI at the beginning of when it first came out, and Steph said something along the lines of isn’t it nice that like you get the ability to like, have that opinion because some of us don’t.

Some of us are disabled, some of us like can’t do this thing. And I was like. Oh, sure. That makes sense. So I guess I’m pro AI now and because she said something like, do you believe Stephen Hawking should be able to use AI to record his thing? And he’s she’s yes.

She said something along the lines of now we’re just arguing about the like, degree of AI use that is okay for [00:06:00] you. We’re not disagreeing about whether AI should be used for publishing. And so I really had a whole turnaround because recently I’ve been having a crisis of writerly pursuits as I’ve become more successful.

I do less of my own writing, just like I mostly editing books. I’m mostly working with co-writers or co-creators. So the words that come out are mainly my ideas, but they’re not necessarily typed by me. And so I’ve been having this whole crisis of if they aren’t words aren’t typed by me, then did I write them?

And somebody said, may have been my wife said something like, if you dictated a book to a typist would they get credit as writer of your book? And I was like, no. That means that the process of typing keys is not integral to the process of writing.

Now if you’re not typing keys there’s a wide range [00:07:00] of things that you can do to be a writer. If James Patterson is a writer, he doesn’t write his own books. If like Rick Rearden is a writer when he’s just presenting a lot of the books is even it being your idea necessary? What is being a writer? And I landed on this idea that like the promise that we’re making with our readers is that we are providing a consistent experience.

If you read my work or this type of work or this pen name, you are going to get this thematic experience, this writing thing, experience, like something that I read and enjoy and think you will like, and that now opens you up to all sorts of other things that you can do. You don’t necessarily have to make a books, you could make a TV show, you could make a board game, you can make an RPG, and you can do all of these other elements.

And I guess I just think that people that are thinking about AI in the way that most people are, [00:08:00] don’t actually hire anyone. Ever. Because the minute that you start hiring people, you lose control over the input. At least you can still control the output of what goes out the door, but you can’t control the word somebody types to someone in customer support.

You can’t control the emails that get sent between your assistant and a blogger or something. You could never control what someone writes back. But in a long way, the more successful you get, the less control you have of what goes out the door in your name. And so all you can do really is control, what that final product looks like.

Not even the iterations of it.

Steph Pajonas: This is so interesting because I was literally having this discussion with somebody earlier today about the fact that story is really what is going to matter the most when it comes down to it in the end because these tools are gonna make other people from being not so [00:09:00] great writers into competent writers. We’re going to get to the point where competent writer is literally the bottom of the barrel. And then anything up from that is gonna be awesome. But then the story is really what’s going to be bringing people into these worlds. Bringing ’em into your business.

You were just talking about all of those books that you have behind you, that is like a whole story. And it doesn’t almost matter the way that it was presented, because you could write books, you could make movies, you could do comic books, you could do so many different things.

But the story is what is really going to be resonating with the people who are consuming it. So at this point, the AI, sure it can churn out those words on the page, but then you make sure that the story is awesome. And then the final product that goes out to people, appeals to them and is one of your best ways of showing off the way that you work.

This is where we’re hitting on things now, finally, in using AI as part of the creative process, [00:10:00] because now we can really work on those stories, those ideas, and make them shine because we have these tools.

Russell Nohelty: Okay, so I’m gonna go probably in a way that no one else would go, which is not every of the books that I’ve ever put out is my best work. Sometimes I put out a book that I think is good, but I was like, Wasn’t great. It had to ship. But I couldn’t spend any more time on it.

And then you put that book out and people respond to it the best of any book you’ve put out and you’re like like the narrative thread there makes sense, but does every chapter have to be like the greatest chapter and do we have to be nitpicky over everything or does it have to hit the right emotional beats. Because okay, I love steamy romance books, but I think everyone can agree that sometimes the words that are written on a page are like vibes. They don’t actually make a whole lot of cogent narrative sense, by themselves.

When you read them. ’cause I do a lot of this [00:11:00] translation to comics and I’m like, you can’t actually you have to look back at the word cloud and how the words are making you feel and the vibes of the thing. Because that is what they’re getting at.

You can’t read Twilight. in the way you read a reference book because it’s vibes. And if some words or pages or chapters are just like vibes, then the cohesion of a chapter isn’t even necessarily necessary to have a great story. You could just be making a leaves of grass crazy thing with weird construction on it.

Stephanie Meyer has to know the coherence of this chapter doesn’t matter. What matters is I need the people to get horny during this chapter. They haven’t been horny in long enough. A chapter could hold story, cohesion but it can also hold emotional cohesion or even emotional discohesion, which I don’t think is a word, but sometimes the [00:12:00] goal of a chapter is that it’s comes out of nowhere and takes you off of your feet and you’re like, what was that?

And the point was that you don’t know. The point of me as a writer is that I know that this chapter was if it elicited the emotional reaction that I wanted, which might be. This doesn’t make any sense. Like, where are we even going with this? These two words don’t make sense.

These two words can’t be together. Then if that is the goal, then the actual writing of the words can be completely disjointed and still contribute to an incredible story.

Danica Favorite: I think you’re right. And I love what you’re saying here because it, it brought to mind two things. First of all, I remember when the Twilight Series came out and a lot of writers I knew in Romance Landia at the time were absolutely infuriated because it was really poorly written.

And by the way, I have not actually read it. I know and I just remember how angry they were that they felt that it wasn’t [00:13:00] great prose. But here’s the thing, great prose are not, it elicited the responses that the readers. Really we’re drawn to, and it was that reader experience that made Twilight such a big hit.

So in the end, ultimately, it doesn’t matter if it’s a great work of literature because that’s not what Stephanie Meyer was trying to do. That is not what you are trying to do with your writing.

One of the things you said about that consistent experience with writers, one of the things when we were talking earlier, you had said that we were curating an experience for people, and I remembered that word specifically because I’m gonna take that and sit with that one for a long time because it really is about what experience are you curating for the readers

Most readers don’t need fine literature. Most readers are looking for that experience. And so I love that point. But the other thing I wanted to say along with that, because this is another thing I’ve [00:14:00] heard you say often about, when you were putting out what you thought was your worst book, being willing to put out the crap anyway.

On your Author Stack, you’ve been talking a lot about that. The idea of it doesn’t matter, just do it. It’s not gonna be perfect. Just get it out there and don’t be afraid.

And so when you look at that fearlessness and not being afraid of, oh my goodness, is my grammar perfect, is this perfect? Is that right? It is the experience I’m creating for my audience, the experience I want them to have. And I think that’s such a different question.

Russell Nohelty: it’s funny, I was talking to a friend of mine who had come to our conference last week and we were talking about how, I’ve had the same editorial team for a long time now. Like 50 plus books that they’ve edited or at least proofread. I was talking about how much better they are now. It was been 10 years they’ve been on books for me almost.

He’s never said anything to me. He’s bought all of my books, [00:15:00] but he’s oh yeah, there were a lot of errors in those books and there are considerably less now and that’s nice. Did it fulfill the purpose and did it teach you to do the next one better?

And if it is horrific, then the fact that nobody cares is great. I used to work in sales and the best and worst thing is that it resets every month. ‘ Cause at the be end of a terrible month, the first of the month you go and you’re like, it all resets. Every single thing resets it’s all from zero again. Then at the same time, it’s man, I had a great month. Oh, March was a great month. April was a terrible month. And at both times, the month ticked over. It started again and nobody knew.

The people that are reading your work have other stuff to worry about. And so like people don’t even know the thing they read.

Hell, I just finished a book and I don’t remember what was in it, what I wrote in [00:16:00] it.

Danica Favorite: I totally get that. There are these different vibes that people have, but it’s not necessarily going to be, oh, I remember this, or I remember that. But as authors, we get so caught up in those few readers who write to say, oh my goodness, you messed up a comma here, or you messed up a comma there. Why are we worrying about those people? ’cause they’re not our audience. Our audience is the people who are seeking that experience and they’re seeking to gain something. So I really loved that story about your editor and how you both are like, wow, we’ve really grown.

Because it is a growth process. And if you wait until you’ve got it all figured out and you’re all perfect at it, then it’s never gonna happen.

Russell Nohelty: It’s so interesting. You’ll like the books that I end up writing with AI or by myself because I know how to write a book and this is the thing that I think gets taken out of the AI debate, both by the way, in [00:17:00] art and writing. If you know what the end product is supposed to look like, either you can write it, you can hire somebody to write it, you can write it with somebody, you can dictate, there’s like a finite number of ways that you can do that.

But if you know the end product that you want and the end product becomes the product that ends up being the thing that it is, is the audience going to care which of these four things that you did. Is Dragon Dictation lesser if the end product is the same? And obviously the answer is no, it’s so ludicrous to even bring that up.

The problem is all of these AI tools can get you 70% of the way there, but if you only know 30% of the craft, you’re not gonna be able to get there. You have to know a hundred percent. You have to be able to know what the output is for any of this AI to work in the way that actually is transformative, [00:18:00] because it cannot work outta the box.

It is ridiculous out of the box. It is laughably bad even at 70%. even if it knows all of the comma splices and everything, like you have to tell it all of the things you want to happen and then you have to yell at it when it doesn’t give you that thing. And then you have to stay on it to not hallucinate and bring in characters who were in other fiction, characters who were like, no, she’s dead.

You literally just wrote that. She died two chapters ago. And they’re like, oh, yes, great thing. And then they kiss your ass and you wanna punch it because I do not need a brown noser. I need someone who knows when characters died in my books to not do that again. Sorry, that was a big

Danica Favorite: No, that’s okay, because I’m gonna tell you something.

I I’m ashamed to say but Steph already knows this. I started plotting a new series with the help of AI. And frankly, when I was redirecting it and saying, no, in this scene, you need to have [00:19:00] this level of sexual tension. Because yes, they are worried about a hostile takeover, but also they’re attracted to each other. And this needs to be both things, which a, as an author only, I know that the AI doesn’t know that. But B, when I told it that it came back and told me I was brilliant and frankly I needed to be told I was brilliant. So

Russell Nohelty: There are some things that I write that I tell it and they’re like, that’s brilliant.

And I’m like, I fucking know. And then there are other things that it’s oh, you’re so right. And I’m, I don’t need you to say thank you Mr. Russell. You’re so smart. I need you to knot it up. Please. That’s all I want you to do is and okay, as I mentioned, 500,000 words into a purely AI written book.

Like an entire AI written series, and it is not an easy series for AI to write. These books are 200,000 words. They’re portal fantasy. They track heaven as if like [00:20:00] it was America over the course of history. And its imperialism as it invades different underworlds and it is dense on a hundred levels.

But it is really pretty good at about 50,000 words. My favorite AI is Claude. Claude can’t go past the a hundred thousand words. So you’re just like, okay, Claude, I guess I’m at Chat GPT now because and I’m I’ve been on chat GPT for several books now, and every chapter, every day I have to remind it. Who is dead, who is live, who is in which portal? Who is here? And I have now two guidebooks. I have a series Bible, and a world compendium, both of which it can reference whenever on top of all the three books.

It’s very frustrating. Writing nonfiction with it is actually brilliant because there doesn’t need to be dialogue and you never have characters just [00:21:00] randomly show up back in visions in the past, romance would probably be pretty easy to write with it because it’s contemporary.

This book series drives me crazy for all of the ways you just said Danica, but mainly because I would like it to be like, do you have to be such a prick right now, Russell, can you be nicer? And I’d be like, I can be. If you’re just going to brown nose me all day every day, it’s just gonna make me madder when you mess up.

Danica Favorite: Obviously Danica and Russell have totally different motivations.

Steph Pajonas: Didn’t they just put out a patch to Chat GPT to make it less of a brown noser? I think they call it Glaze or something like this. This is the glaze of GBT. Oh. It’s like P

Russell Nohelty: glaze or whatever. Glazing

Steph Pajonas: it.

Danica Favorite: Yeah.

Steph Pajonas: Yes.

Danica Favorite: But could we just 500,000 words you guys like, this is what Russell has written fiction with AI.

500,000 words.

Steph Pajonas: Yeah. And let’s go back to your point about the crafts, that the craft of [00:22:00] storytelling is important. You can get these 500,000 words out of the AI and get it to give you some quality stuff because you understand the craft of storytelling. And the people who don’t and try to come in and make stories of some, maybe one or two, percent of those people are gonna get lucky and make something really great.

But then the rest are just gonna be like, why doesn’t this make sense? I don’t understand why nobody likes this story because they didn’t understand the craft.

Russell Nohelty: Yes. And honestly, because I have lived through the era of content farms, I can already, for almost no money, get a book that is bad or serviceable or even good. People have this thing of oh, AI’s gonna throw the internet into chaos. Although there’s been unlimited crap on the internet forever and they could filter that out.

They choose not to. It’s not because [00:23:00] AI is this existential threat I don’t think to social media. I think it’s because the tech companies that are invested in AI in a way they were not to content farms. So they want the slop to happen so that it can get trained better and so they’re letting it happen, but they don’t have to let it happen.

Amazon has no incentive to surface crap books because they want people to buy things, not just experience the platform for longer. And so while the internet world of like engagement may be horrific with AI slop. I think that because people have to actually read the books and actively open the thing to read it, that Amazon is incentivized to bury that just like they buried the AI slop from content farms. And so I think that for the [00:24:00] author world specifically, like the book world, like not the, I write tweets, not the AI art world, which is very different than and I think is going through a existential crisis because it’s one image that you were consuming.

Anytime there’s one image that you were consuming and that is the most that you’re getting, AI has the opportunity to cause serious harm and existential stuff to your business, but for books specifically and for things we are packaging for money specifically and for things that carry our name specifically, where people are going to reach out and say, I want this person’s book.

There is no incentive to put out bad books. And whether you use AI or you put yourself, there is a chance you’ll put out terrible books. Like you can write terrible books very easily.

But like for us as authors or [00:25:00] people that make comics or people that are telling stories where a story narrative needs to be consistent across all of these, all of the things that you put out where your name is on it. We have a very, I think, privileged position in AI, which is we can use AI in a way that furthers our narrative because we know what we’re doing. Without as much worry that a bunch of garbage is gonna come in and drown it because Amazon and book retailers are historically excellent at filtering out garbage because they are only incentivized when you spend money.

Steph Pajonas: Goodness. I’ve been saying this for two years now, like people who have been worried about the tsunami of crap, right? That we heard about from Joe Konrath, like at least a decade ago. They’ve all been saying, what about all these AI books that are gonna come and [00:26:00] flood the Amazon store?

And then I’ll never be able to be surfaced and read by people. And I keep telling them like maybe they’re gonna have an impact for a few hours. For a few hours and then Amazon’s their algorithm is just going to surface the good stuff and it’s gonna push the other stuff down into oblivion, which is what happens to a lot of people,

Russell Nohelty: right.

That write good books too. Like Amazon is more likely to not surface good books just because their algorithm is filtering out all of this. The fact of you not being read is a problem.

Steph Pajonas: Sure.

Russell Nohelty: Be like, even if there were zero crap books, you would still have a problem getting read.

Steph Pajonas: Absolutely.

Danica Favorite: Yeah. Yeah. And that’s a completely different problem than the actual AI written books.

So I’m curious because I know you do both fiction and nonfiction with AI and you’ve talked a little bit. Could you tell us your workflow, what does your workflow look like? And I know you were saying that the non-fiction is really different from the fiction, so if you could just tell us something about [00:27:00] each one and what that process looks like, because I think that is always what’s fascinating to our listeners and we’d love to hear that.

Russell Nohelty: Well, I did a really good course on the non-fiction stuff for Future Fiction Academy. I don’t know if you guys have any association with them or know them at all, but like you should, you could go and figure that out for. Okay. So I have three main custom GPTs. One is trained on viral posts for social media, I don’t use that very much anymore. Even if I’m just copy and pasting. I just can’t stand doing social media anymore.

And then I have one trained on all of my old posts and my books. And I have a lot of them. I have written or co-written 16 nonfiction and just hundreds of posts that I’ve trained it on with most of my major concepts.

And it’s actually very good now, 90% of the time mimicking my voice sometimes it makes [00:28:00] me seem like a right asshole. Do I really sound like that Chat GPT? Really? That’s so mean. I’ve never said anything like that in my books. And they’re like, you’re right. I don’t know if I’m right, but this is the level of brown nosing that I’m okay with. I’m okay with you saying, no, I’m not like that.

But so it’s quite good. I have used it for everything. Somebody wrote a chapter in a book that I need to now strip that chapter and have it rewrite it. I will go and I will say fill the space. This chapter is now not in your memory. Take this out of your memory. Take this out of this book and fill this space with the thing.

I have taken our courses like Story Urge and Story Urge became How to Write Irresistible Books That Readers Devour. God, that’s so long. And basically I just said take all of the transcripts and make this chaos make sense. Because the thing about Monica and [00:29:00] my combined live courses is they’re great in the moment, but you can’t go back and listen to them. It’s chaos. I couldn’t go back and make notes of it. So I took the transcripts and I went and brought all of them into Chat GPT and then I edited the heck out of it.

I have a book coming out based on all of the transcripts to Lee and my Six Figure Author Experiment where I did the same thing. These are the elements that I want. It’s called the Six Figure Author Blueprint. What are the main themes? How do we make this actually make sense as an integrated author ecosystem?

It’s really hard to have a blog post, less than a thousand words, and if anyone take it seriously. But most social media posts are about 300 words. And so I will often develop a concept that does really well on social media, and I will say, take this and make it a thousand words, because it’s just filler. I did the work. I built the concept, like the framework is here, it just needs to be three times as long. So I use it when we have a [00:30:00] podcast episode to do story notes. Like those are the normal ways.

My favorite way recently is I’ve been using deep research to take books that I’ve written and create workbooks from them. You know that real neat workbook that I gave you at my conference, Danica?

Danica Favorite: Oh, I know. I was really hoping you would talk about the workbooks because it blew me away that we would sit there and we were talking, you’re like, yeah, I just made another workbook last night.

I’m like, what? Yeah, I’m I make them a workbook queen. I’m a sucker for workbooks. Yeah. So I wanna hear about your workbook process here.

Russell Nohelty: The book I gave you, the one for this campaign that’s live right now was AI made. AI made it and I then futzed with it. But I put it into deep research and it’s a horribly named product because maybe the worst thing that it does is research.

What it does really well though is take the concepts of five books or 20 things or one book and make an RPG from it. Something that is complex [00:31:00] and transformative and takes things and makes connections between them in ways that you wouldn’t even notice.

In the capitalism workbook, there is a module for soul resonance statement. And I was like, oh my God, I didn’t, that’s not in the book. Where did that come from? And then I thought about it and I was like, yes. This is a necessary addition to get from step one to step three. You need this in step two.

And the rest of AI, unless you’re using Plotdrive, is pretty well capped at a thousand good words. It will say, this is a 2000 word article, but all it will do is make shorter sentences and make it longer. So it looks like the length of a 2000 word article, but it’s really only 800 words.

And I’ve now tested this, I’ve written 500,000 words and I cannot find a way outside of using Plotdrive or Raptor Write of getting above a thousand words. It will tell me that it figured it out, update like so many times and it will figure it out for [00:32:00] 30 minutes and then it will always slide back down.

But Deep research does not have the same problem. And I know you’re probably gonna be like, use Gemini or Claude or some other thing, but I’m now deep into ChatGPT. But so what deep research has allowed me does is like it will give me a hundred page document back. Or 50 page document, it’ll give me a length of document that is appropriate for the transformation, whereas most other ways that you input information will not give you that length back.

I’ve used it to make world bibles. I’ve used it to make RPGs, companion books. With fiction, things I love about writing and there’s things I don’t like. One thing I don’t like is I like that tedious backstory exists, but I don’t like writing it.

So there are things like, I need 12 scenes of this thing going from X to Y. Go. It’s great at that. Make 12 bad guys to [00:33:00] fight and make it harder each time and make them all unique. I don’t wanna do any of that work. I want that work to have existed. So that’s a really good way that I use for like my fiction. I don’t particularly like purple prose, but fantasy fans do.

So I’m like, add all of the things, make it lush and rich and I’m just wanna write. He kicked her butt. Then you make it a chapter, whatever. I don’t care.

Which is what I do in comics, by the way. In comics, if there’s just a fight scene, I will tell my artists, make a thing kicky, punchy. Sexy times, make it happen. I don’t need words, I just need actions. And so whether it’s my artist coming back or AI coming back, or a co-writer coming back, there are just things that I do not like.

Things I do, like splitting people into groups and putting them through hell. It is not great at that. And it’s not great at remembering what groups are, where is in [00:34:00] which group.

I have to be on it all of the time. But it’s really good at telling me where an arc should go or like letting me talk through an arc when I will wake up and I’m just tired. So like I’ll be like, okay, I need three to five chapters about this.

Then it will gimme something terrible and then I’ll be like, actually, what about this? And I end up riffing pretty well. I actually think the riffing is one of the more underutilized things that these things can do. I think people like to tell a thing what to do. Then have that thing come back. But that is actually a very banal way, which AI can help you. AI is much better at knowing what you actually want to do. It opens up more possibilities and you will get a more interesting product. But again, you have to know the constraints to put on it.

I don’t know, does that answer any of your questions?

Danica Favorite: It does. I love that. I think it’s great. And actually, I love that you brought up riffing, because that is my favorite thing to do. When I was plotting this book thing, I have [00:35:00] 55,000 words of just me and the AI riffing, just us saying, okay, what about this?

Okay, no, what about this? Oh no, I like that, but change this. And that to me is the fun in the creative process. But I’ve always been really into brainstorming and having fun. Hearing your process and hearing what other people like until you mentioned that about the riffing I had not even thought of that being one cool thing to do with the AI.

And I’m like, yeah, I have 55,000 words to prove that might be my thing.

Russell Nohelty: I spend a lot of time being like, okay they’re supposed to fight, but like, how can I make it so they actually play Parcheesi instead?

I spend an inordinate amount of time trying to make people not fight for someone who writes action adventure fantasy.

Danica Favorite: I think that’s fun. And so do you feel like with your writing process and the different workflows that you do how much is that different between your fiction and your nonfiction?

Russell Nohelty: Well, I’m a lot nicer to AI am my nonfiction because it does the thing that I want it to do. I [00:36:00] can’t get over how angry I am when it forgets a thing we’ve been talking about in this arc. In the arc we’re currently writing. I for some reason think that it should remember. All of the context that we wrote at least that day, if not the things that it said, it can hold in its memory, it should be able to remember that.

I feel like I have less constraints on it with nonfiction because what I’m trying to do is figure out the best way to demonstrate a concept. And often the best, most transformative way to demonstrate a concept is not the way that you think it is. Whereas with fiction, everything has to follow from the thing that happened before.

So the less constrained you can be, the better. I think the more cohesion you can have with the AI. Because it’s not at odds. You allow it to riff. [00:37:00] Whereas the longer a book goes, the less you can let it riff because it breaks the world. Whereas if you set something in real world in capitalism, it has to flow in a way.

There are ways to make Shopify work differently than Kickstarter, but at the end we’re talking about buyer psychology. There’s a finite number of ways to demonstrate how to write a sales email. And Chat GPT probably knows all of them. Like it knows probably better than you how to demonstrate that concept because everyone else has been trained on it, but it probably doesn’t know the gravitational functionality of a portal in a world that you just constructed based on a world that came out of your own mind.

Or even how to get these two [00:38:00] characters actually horny for each other. Because it’s more than just like smashy smashy together. There’s a nuance that AI can’t have because getting someone to sex or to fight is unique for every book that will ever be written. Because the two characters are unique and they have different turnons, they have different things that make them want to fight.

They have different situations. And so the more unique a situation is, the less likely the AI is to see it, have seen it before, and then the more likely it will either pull it back to something you don’t want or do something completely unexpected, but not in a good way.

So nonfiction is nice because you can be like, I’m writing a book [00:39:00] about how to thrive as a writer and a capitalist dystopia. Who are the 10 best writers on this topic? What are the things that other people say? How are some other ways to demonstrate this concept given this context? And because we have had people complaining about capitalism since the day after capitalism existed. There is abundant research that I don’t have access to that it can pull from. And it can say, actually I think this way is better because of X, Y, and Z reason.

Even with something like the author ecosystem. There’s five ecosystems. They work in a way, and they work that way every single time. And so I can tell it, go make me a forced situation given these contexts, and it’ll probably be pretty good.

But I’m trying to demonstrate the mean, and as long as you want AI to [00:40:00] demonstrate the mean or even the median or even the mode of somebody’s experience, it is really good at simulating examples. It is really good at taking your concept and then beefing it up or telling you where there are holes in it or telling you where other things are.

Nonfiction is almost always, maybe at best, it’s taking a novel concept and then explaining it through very known variables.

Danica Favorite: I think that just beautifully illustrates exactly why AI isn’t gonna take writers’ jobs anytime soon and why it really is the writer’s skill through the AI that makes that work transformative.

Russell Nohelty: Yeah. I don’t think LLMs will ever be able to take writer’s jobs because of the way that LLMs specifically operate.

There might be a different [00:41:00] AI, like organic brain AI or there might be an AI system that is better at taking all of this stuff and synthesizing it together that might effectively be able to make fiction as well as us, or even nonfiction as well as us. But it is not gonna be LLMs.

Danica Favorite: Yeah. Absolutely. So last question. Tell us about your favorite AI tool. It sounds like Chat GPT, but do we have a winner elsewhere? Is there something else you really love?

Russell Nohelty: I’ve been thinking about this a lot. I think it has to be Chat GPT because I use it so much. But I feel like Google search is also a really great AI tool and it’s been there forever and Substack’s algorithm is a pretty good AI tool as well. And I like those ones.

If you’re talking about specifically like the kind of AI that can write books, then it has to be, actually, I like Claude better. The problem [00:42:00] is Claude does not allow you to upload enough document length things for it to work for a book that is longer than a hundred thousand words currently.

Claude has HIPAA compliance if you are in healthcare, like there’s a HIPAA compliant one. My wife works in healthcare, so it’s really good for that. But I will say, ’cause I can’t think of a one that I like better than Chat GPT now is Chat GPT.

Danica Favorite: Awesome. I love it. I love it. Russell, thank you so much for coming and hanging out with us today and telling us about all the cool AI things because I know that this was a lot and I think this is one that I’m definitely gonna go back and listen to a few times ’cause there were a lot of great tips and tidbits in there and I think our listeners are really gonna relate to it as well.

So thank you so much for coming today and I’m really looking forward to seeing all of the new cool things that you keep doing with the AI. So thank you.

Russell Nohelty: Thank you so much for having me. I’ll talk to you soon.

Steph Pajonas: Absolutely. We’re so [00:43:00] excited that you came. So we’re gonna make sure that everybody gets all the information that you dropped on us in this episode.

I’m gonna make sure that we pull the transcript and put together an awesome blog post for people to come check out. It’ll have all the information in there and anything that Russell mentioned, including links to his awesome Author Stack so that we can send some people to you for some more awesome innovative thinking. That’s what I like to think of. Russell is very good, innovative thinking.

All right everybody, so come on by to brave new bookshelf.com, read the show notes. Click on our links and as usual, Danica would tell you that we would love for you to like and subscribe. Right?

Danica Favorite: Find all the stuff for Publish Drive, find all the stuff for Future Fiction Academy. Give us those likes, give us those subscribes and share the love.

Steph Pajonas: Absolutely. All right, so from all of us, we’ll see you guys next week, so bye-bye bye.

Thanks for joining us on The Brave New Bookshelf. Be sure to like and subscribe to us on [00:44:00] 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.

Episode 42 – Innovative AI Storytelling with Russell Nohelty from Author Stack