30 - Legalities, Workflows, and Transformative Use with Alicia Wright

In this episode of Brave New Bookshelf, we’re joined by Alicia Wright, an IP attorney and science fiction author, to explore the legalities of AI in publishing and how authors can ethically and creatively use AI tools in their workflows. Alicia shares her expertise on copyright law, explains common myths about plagiarism and AI, and offers invaluable tips for integrating generative models like Claude 3.5 Sonnet, GPT O1, and Novelcrafter into your writing process. Whether you’re curious about fair use or looking to streamline your storytelling, this episode is packed with actionable insights. Visit our website https://bravenewbookshelf.com to view the full episode notes, links and apps mentioned in the episode, and the full transcript.

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

**Steph Pajonas:** Hello everyone and welcome to the Brave New Bookshelf. I'm Steph Pajonas, CTO of the Future of Fiction Academy, where we teach authors how to use AI in any part of their business. We're just trucking through January here. So there's always the jokes out there that January is like the longest month ever, right?

It feels that way. And we're just trying to make it through the month and on to warmer, greener pastures, hopefully in the springtime, right? And I'm here, of course, here with my lovely co host, Danica Favorite.

Danica is in, she's probably got more snow than I have. I don't know. Probably. 

**Danica Favorite:** How much snow do you have? 

**Steph Pajonas:** I have about like two inches outside still, but it is [00:01:00] melting. It's melting. 

**Danica Favorite:** Yeah. That's about what we have. So we're the same. All right. So I'm Danica Favorite. I am the community manager at PublishDrive, where we have a whole ecosystem for authors to help them on the various stages of their publishing journey, whether that is formatting their book, crafting perfect metadata and book descriptions, to doing their book covers, or distributing their books, or then once they get that book published, distributing it to the widest market possible.

We've got you covered through all areas, and then even once the book has been published, if you need to split royalties, we can help with that too. So, here we are, helping authors with all kinds of things, and as we were talking a little bit before we got started today, Also making sure that one of the areas where we're helping authors is educating them about what's happening with AI, and what AI looks like, and we don't care whether or not you use AI, but as we've discussed, it's going to be pretty inevitable, but at least you have the information to make [00:02:00] wise decisions that are best for you and your author business. 

And I am really excited about today's guest. Our guest today is Alicia Wright. She is amazing. I'm just going to say that. I have been in awe of Alicia for a long time because she is just such an incredible person and really has a lot of knowledge about AI, particularly in the legality department, because I think a lot of us we hear people say, Oh, this is legal.

This isn't legal, et cetera. And of course Steph and I, the FFA and PublishDrive, we all take the stance that AI needs to be used legally and ethically. And Alicia is an expert in that. So, Alicia, I'm going to let you introduce yourself, tell us a little bit about yourself, your areas of expertise, and why we should be listening to you.

Because, I know why, because Number one, you're super smart, [00:03:00] but let's let's take it over to you. 

**Alicia Wright:** Thanks so much. That was a incredible introduction. As Danica said, I'm Alicia Wright. I am an IP attorney with a computer science background. I have a bachelor of science in computer science, a master's in computer science.

When I decided I didn't want to practice in the field of computer science, I went to law school. I got my JD, which is a law degree in the U. S. And I've been practicing as an IP attorney for 17 years, specializing in patent law, where I see a lot of AI related applications. When I worked at a law firm previously, I was considered one of their AI specialists.

Now, because I am a writer. I've been writing for as long as I've been practicing law. I've also made sure that I stay involved in copyright law 'cause that's very relevant. So I've touched the AI technology a lot. I've written over a hundred patent applications related [00:04:00] to AI inventions and I've been involved in copyright law as well.

So when AI went mainstream in the writing industry a few years ago. That was really exciting. I was initially thrilled because I've been seeing patent applications for generative AI for about a decade at that point. And then it was really a bummer when the writing industry wasn't as open to it as I was, but I'm excited to talk to you guys today about AI and the legal side of that.

**Danica Favorite:** Yeah, I love that. I think this is so great because, as we were talking a little bit about before the show started, how, there are all these memes and people are afraid of, Oh no, this is illegal. So much of that is misinformation. I really love that we have an actual lawyer who knows about IP and AI and all of that, actually giving information.

And so, obviously, we do want to caution you, Alicia is not your lawyer. So, check with your lawyer if, there's any [00:05:00] specific legal advice you need. This is really just a general background and overview on AI and AI legalities.

I want to go back to something that you said. You were seeing these patent applications for generative AI 10 years before it came on the mainstream.

To me, that is just amazing because all of the authors and author community are just like, Ooh, it's this new thing. We're so scared. But you've been seeing this legally for quite some time now. 

**Alicia Wright:** Definitely. I've been drafting patent applications, mostly related to image generation, a lot of it facial recognition and generating fake faces, patent applications related to that for, yeah, a decade ago.

It's been a while, just part of my career. So when it went mainstream, I was like, it's here. I can touch it. I can use it. This is so cool. 

**Danica Favorite:** Yeah, that's that is very cool. And I love that, you've gotten to see that beginning of the technology and how it's evolved. And so now, you know that, Ooh, I can't wait, I can't wait.

And now [00:06:00] it's finally here, you can use it. So let's talk about that. How are you approaching AI and publishing? And how has this coming on to the mainstream, why is that exciting to you? And what are you doing with that? 

**Alicia Wright:** Well, it's exciting to me because I love technology. I've been a math science person longer than I've been a writer.

So just being able to use technologies that I learned about years ago is super cool because when you're writing a patent application, the inventors haven't even necessarily put it to practice yet. So most of the time, I don't actually get to use it and get to see it, but I still have to understand how it works and be able to describe how it works.

So it's a different thing to be able to get my hands on it, which is really cool. But how am I using it? I use AI a lot, especially on the brainstorming side. I think of A I as a partner and very much brainstorming partner. Tell it the ideas that I have. It helps me connect threads together. I say I want to do this and I want to do this.

And how do I make these things [00:07:00] work together? And usually I'll ask it for a bunch of ideas. Give me 10 ideas to do this and maybe one of them will be great. Or maybe all of them will be terrible. But I will look at those and think, why doesn't this work for me? And that will help me figure out the connections that I wanted to make.

So from brainstorming, we go to outlining. I use AI a lot in the outlining phase. Once we have the ideas together, how to generate the first draft of my Save the Cat outline, because that's where I start. We make a lot of changes to that. I'm guiding the process of changing and then we blow it up beat by beat.

I want to do 10 scenes for the fun and games beat of Save the Cat. And I want the scenes to include, maybe I know what some of the scenes are and I don't know what the other scenes are. So I ask it to give me ideas. I actually usually ask it to give me multiple sets of ideas. Like give me 10 scenes in a row and give me five sets of that and look through, what's [00:08:00] common, what's necessary in all of these and piece them together.

And this is similar to how I did it before AI is that I would say, I want to get from here to here and I would brainstorm quick the scenes I need and then flip the page and brainstorm again the scenes I need and flip the page and then go through them and think, What is in common? What did I keep doing over and over again?

Because that's something that needs to be in there. So I do that except with AI. So my process faster and I love that. When we get to actually writing. I tend to dictate and then have AI clean up the dictation. I've experimented with having it write for me with fine tunes with telling it my writing style, and I can actually get it to imitate my writing style quite well more lately as it's getting better and better, but I'm just so particular.

I end up editing it to death. Even if it's in my writing style, that's not how I would have done it. So I just do it myself. And then [00:09:00] have it developmental edit scene by scene. I usually have it output a log after each chapter so that it can go back and view the log again, because it's not going to remember the whole conversation.

So we're basically working from a log as we go through each chapter. So that's, I think that's a bit, that's it. That's, that's how I use AI. 

**Steph Pajonas:** Wow, that's a great idea. The log. I'm stealing that one. I'm stealing that one. 

**Danica Favorite:** I need, I need the prompt for this log. 

**Steph Pajonas:** Make a log of all of your changes and tell me what happened in this chapter.

That'd be good. 

**Alicia Wright:** Track characters and character development and setting so that it's. Things are consistent. Actually think of it more. It's like a beta read where it keeps a log chapter to chapter. 

**Steph Pajonas:** I love it. Yeah, so you got you use it as more like a helper.

It's there to assist you along the way. And I've heard you say this, and I've said it a few times. It feels like a junior writer. You're asking it to do some things. But then you are like the [00:10:00] ultimate director and you make sure that everything is in your voice and the story has gone the way that you wanted it to go. 

**Alicia Wright:** Absolutely. I am fully confident that at the end of the day, what comes out is me. That is my authorial voice. And I had a junior partner that helped me along and we did it faster. And that's fantastic. 

**Danica Favorite:** Yeah. So since we're talking about, like your process and copyright law, how does this impact your copyright?

Like, is this work with this workflow that you just described, is that copyrightable? 

**Alicia Wright:** Absolutely. So the brainstorming, all that is considered ideas, and we don't even get into copyright eligibility for that. But the writing I've dictated, so all of those are my words. The fact that AI touched it and edited it and tweaked it and cleaning up the dictation doesn't change the fact that I authored it.

Like AI plus me still includes me and my part of it is copyright eligible. And once you've worked together with AI such that your work is [00:11:00] integrated and entangled with AI's work, you can't separate that. So the U. S. Copyright Office would say that you have limited copyright protection in that what is yours, what you contributed to it is copyright protected, and what AI contributed is not.

However, the things are so entangled that if you take a page and copy it, you have copied my work, and that is copyright protected. 

**Danica Favorite:** That is really good to know. And I think that is so important for people to understand that, it's that enmeshing that really makes it your work, because I think that a lot of people that that really is the question.

I think going around. Is this your work? And I love how, as you've described your process. And talked about that, it's really clear. This is Alicia's work. This isn't the AI doing something for you. So I love that explanation because I think that for a lot of people, I know Steph and I see this in the group all the time where people are asking, well, it brainstormed [00:12:00] this thing for me.

Is it really mine? Well, yeah, all three of us right now could have the same exact idea coming from AI and we could write a story. Even though the idea came from AI, the three of us are going to write three completely different stories based on that idea. So, 

**Steph Pajonas:** And it's the expression of the idea that's copyrightable.

It's not the idea itself. It's the exact expression, the words on the page. Those are the things that are, the copyrighted part of the process. 

**Alicia Wright:** Right. And it can really become clear if you think of if everything that's being done by AI was done by another person. No one would question that I am the author of that.

The other person is also the author, but no one would question that I put work into that, that I am the author of that. That my work is protected in there. And just because the other person is AI who can't own anything, that doesn't take away the fact that, that it's my work. 

**Steph Pajonas:** Yeah, that's true as well.

So it becomes like a co writer, And even though it's part [00:13:00] of the process is not protected, you're part of the process is. So, it becomes enmeshed. And then do I pick out the little words, like I used the word the and emperor, and Fantastic in this sentence, but then AI put it in these other ones.

How are you even going to know? 

**Alicia Wright:** Yeah, for all practical purposes, it's just, it's protected. You can't take that and, and use it and say, it's not Steph's work. 

**Danica Favorite:** I think that is really good. And I love that again, we're having these conversations because there's so much misinformation out there. And so, for people who are, looking at things on the internet and saying, Oh, well, I heard this. Okay. Who is your source? Where did you hear it from? 

Obviously we have an IP attorney here talking to us about that, which is really important to understand that when you're like, yeah, but my neighbor, Bob said... Okay, what's your neighbor Bob's experience here.

Do your due diligence. If, you're [00:14:00] writing with AI and you're worried about copyright? Dig into some of the research. We have some great resources in our group. Obviously, Alicia is a wonderful resource. 

Alicia, what are some other things authors should be wary of when they're worried about copyright or they're worried about, making sure that they're following whatever rules need to be followed when they're working with AI. 

**Alicia Wright:** I know some authors are concerned that if they're using AI is what comes out of AI somebody else's is it somebody else's copyright protected work that AI output and can they use it. 

And to that, I would say that these generative AI models are huge statistical models that were trained on billions of other texts, and the chances that you've asked it a very specific question related to your very specific idea and its output someone else's word for word is so incredibly small that I feel safe saying it's never going [00:15:00] to happen to me. 

**Danica Favorite:** I think that's really good because I think that is the main thing. Even now, with our AI survey that we did, I still got in the comments, "All AI is plagiarism." And I'm like, no, you don't understand AI. So I love that you're addressing that because just based on what you've said here with the models, and this is something Steph and I have said as well, is that no, it's not plagiarism.

It can't be because of just how random it is. 

**Alicia Wright:** And in some of the pending lawsuits that artists and authors have filed against these AI companies there have been motions to dismiss from the AI companies and the courts have generally dismissed the part of the complaint that the generated output Is there a copyrighted work they dismissed that part of the complaint because it's generally not.

**Danica Favorite:** Yeah, so that is actually a really good point and I would like you to explain this because not everyone understands what does it mean when they dismiss part of a complaint. In layman's terms how [00:16:00] should we say oh they dismiss this. What does that mean for us. 

**Alicia Wright:** It means that the plaintiffs haven't been able to show enough evidence to move forward for that part of the complaint to be heard.

So generally with these copyright cases, they're claiming there are really two instances of copyright infringement. The first is that you took our copyrighted work and used it to train your AI model. That's instance one. The second one is the AI model outputs our copyrighted work. That's instance two.

Instance two is when heard by the courts. As I've seen so far is being dismissed because it's not outputting that we can't show that what's being output is their copyrighted work. 

**Danica Favorite:** Yeah. So essentially, they don't have enough evidence to prove that it's that the output is copyrighted work.

**Alicia Wright:** Right. And also to be clear on that first instance, it still hasn't been proven that copyright infringement happened because the fair use analysis has not been considered yet, but they haven't been able to dismiss it [00:17:00] because it has been shown or it can be shown that their copyrighted works were copied to train the AI so that those parts of the case are going to proceed.

That doesn't mean there's actually copyright infringement. It just means that the court can't throw it out. They have to consider it. 

**Danica Favorite:** Most of us are writers without a legal background. So when people say, Oh, this part was dismissed, or this happened, or this is going on, it just is really good to have that background of what that actually means.

So thank you for those explanations. 

**Alicia Wright:** You're welcome. 

**Steph Pajonas:** I think I wanted to touch back on the whole oh, if I prompt it, am I gonna get plagiarized works out? Because then there's also the flip side of that where people fear that because they've been using AI to brainstorm and they've given AI all of their ideas and even had it draft some of their works, that that is going to then go into the training data and then show up in somebody else's work in the future.

And I [00:18:00] have to reiterate that this is also statistically pretty improbable, like very, very, very, very tiny amount, and it's not worth worrying about. So it's like the two sides of the plagiarism coin that I hear from people. And we're just here to reassure you that I don't think that this is something you need to worry about.

**Alicia Wright:** So there have been instances, and especially when AI first went mainstream in the writing industry, folks who were against AI were making a point to surface instances where AI outputted word for word a well known text like the Declaration of Independence, for example.

And it's important to note that the texts that AI was able to output word for word are very famous texts. They're texts that appeared multiple times in the training data and the person sitting at the computer who prompted that output was probably specifically prompting in order to get that text word for [00:19:00] word.

So this is a very specific situation. It is not what generative AI is designed to do. 

**Steph Pajonas:** Correct. If you're going to go ask for Dickens or Jane Austen, and you're going to ask for specific passages, you're probably going to be able to get some of them. But it's not one of those things where you're prompting for your own, future sci fi work that involves robots and androids, and then you suddenly get stuff from Jane Austen. It's just not gonna happen. It's that is statistically improbable. 

**Danica Favorite:** And let's be honest, yes, you could prompt the AI and say, give me some lines from Dickens or this famous book or whatever. You can get the AI to do that, but it's just the same. If I went to one of these books on my bookshelf that's not mine and copied a passage from it. It's the same thing. It's just, you're just asking the AI to do it rather than doing it manually. 

But that isn't how most authors are using AI. And the [00:20:00] only way that you're going to get that kind of response is if you're doing it very deliberately and very intentionally, just as deliberately as if, like I said, I was to grab one of the books off my bookshelf and copy from it.

And that isn't what generally the AI is doing. 

**Alicia Wright:** Right, because these companies didn't spend billions of dollars to create machines that can output text word for word. That would be a waste of money. We can do that already. We can grab the text ourselves and copy them. That's not what they're for.

They're actually designed not to do that because that would not be a useful use of their money. 

**Danica Favorite:** Yeah. I love that point. I think that's a good one that, it's not a good use of money to get it to do something like that. Like the whole point of the development of AI is that additional creativity.

So I love that. Thank you for sharing that. 

We've talked a little bit about this already where you were talking a little bit how you're using AI. So let's go into what your workflow looks like. Do you have a particular workflow that you're using with AI that feels good or that you [00:21:00] find really is giving you some good consistent results?

**Alicia Wright:** I use NovelCrafter. Even though I don't have it generate full scenes for me. I do have it helping with transition sometimes, but I love the codex in Novelcrafter, and I generally dictate in to Novelcrafter and then have a prompt where I copy everything I dictated in the prompt will just generate the fixed text on top of it.

And I go through scene by scene like that. If there are gaps, things that I want to fill out, because sometimes when I'm dictating, I say, describe this house. Cause I'm, I'm not good with that. Then I'll go back through and copy this part and Novelcrafter codex will kick in and I will ask you to describe the house.

And then I rewrite that because I didn't love it, but it spurred my creativity and reminded me what the house actually looks like from the last time I mentioned it. So that, that's my workflow. I go scene by scene. I tend to edit [00:22:00] immediately after I write and then move forward just because if I do the whole dirty first draft thing, then I have to fix it.

And that's the worst part for me. I used to find that I stall on the second draft. So I fix that by editing as I go. You have to know yourself. 

**Danica Favorite:** I'm really glad actually that we went a little deeper into your process because I did not know you could dictate directly into NovelCrafter because I also am a dictator.

I really related to you earlier when you were talking about how you've tried the different writing things and, and really dictation still is what feels best to you because that's exactly me. I have a much longer dictation process.

And I'll be honest, I still have not figured out NovelCrafter. I really want to love it. And I really want to use the codex because everyone that I know who uses it and loves it says it's mind blowing, I just haven't figured it out. But now I'm like, wait a second, [00:23:00] if I could dictate directly into NovelCrafter, I wonder how that would work.

**Steph Pajonas:** it's a browser based tool, correct? Like, that's, that's how I remember it. The cool part about the phones now is that they're just so good at understanding spoken words. So, the dictating into your phone, it's like I dictate text messages all the time.

So I can easily open up a voice memo and also just brain dump out my story whenever I want to, and then that gets transcribed, but then you could do the same also with NovelCrafter. Like she's saying, she could just talk into her phone and have it right onto the page and then boom, that's all you really have to do.

Highlight, run your prompts. I love 

**Danica Favorite:** that. I think that is really exciting because for example, I've seen a couple people demoing Notion where they've got the dictate button and you can dictate into Notion, but not everyone has it yet. And I'm one of the people that doesn't have it yet. And I'm Notion people, if you're listening, I want that button, please.

Like, [00:24:00] yeah, using, using, using the podcast for my own selfish aims. But yeah, like, I do think that's a super cool point, Steph, that like, now, there are so many more things that we can just dictate and... 

**Steph Pajonas:** Dictation is really still very popular. Kevin J. Anderson was just on the Creative Penn , talking about how he still dictates. He still does the first drafts of everything. And there was a time when he would have to send it out to a service that would type it up for him, and now he's able to actually use AI to do a lot of those first drafts of stuff because it can go over his dictation fairly easily. 

And I like to walk. I walk a lot. And so sometimes if I'm, my brain is everywhere, I'm thinking about a story, I just get out my phone and I dictate into the voice memos. And it's very disjointed. It's just me saying. Saying stuff about my stories. And then I dump that into Gemini and I'm like, okay, here's the last time I talked about this [00:25:00] story.

Pull everything out that we need to talk about. And it does it for me. So that's great. 

**Danica Favorite:** And that's part of my process too. Because like you, I haven't lately, when you were talking at the intro about the weather. I'm telling you, we've had record cold here in Colorado and it's been miserable and I have not been going for my walks.

Unlike you, I don't have a dog that has to be walked every day. So, but yeah, I, I love, I do that too, and as I've mentioned before, my one hour each way drive to the chiropractor that's dictation time for me. So, I love that Alicia is just like, yeah, I can dictate right into NovelCrafter. Well, maybe I will give it another try. 

**Alicia Wright:** I was so reluctant to start using Novelcrafter, but there are so many little things that I love about it. Like you said, Steph, I love that it's browser based, which is odd for me because I'm typically like I want my data to be protected. I don't care so much that it's, In a cloud, I'm not worried so much about people stealing my data.

I generally want it on my hard drive [00:26:00] with my external backup. So I know that it's backed up. So working with things stored in the cloud is, is different for me. It made me reluctant to try Novelcrafter, but I love everything about it. 

**Danica Favorite:** This is my little plug for Plotdrive. I love that that is also a browser based tool that the FFA in conjunction with Realm Chef have come up with because same as you, Alicia, I am very reluctant sometimes to have everything cloud based and all of that because, in case something fails, I'm a little paranoid writer here, but yeah, so I think that may be my next frontier of exploration is, these different cloud based apps where you can dictate directly into them.

I'm curious, Alicia, with you and your dictation process.

Like, do you ever just find when you actually have to sit at the computer and actually physically type? You're like, wait, what? 

**Alicia Wright:** That has not happened to me yet. Probably because, my day job as a lawyer, I'm also at a computer and I don't dictate. So I'm [00:27:00] still just as attached to typing.

**Danica Favorite:** Yeah, it's like, at least for writing wise, like I'm the same way, like for work, I can like, but. I'm telling you, like this weekend, I sat down to try to write and I was like, Oh, I don't want to type. Oh, this is hard. Oh, I'm going to clean my house instead. 

**Steph Pajonas:** Cleaning is procrastination. It happens, right? 

**Danica Favorite:** It is. It is. So, we've talked a lot about your different tools and things to use. Do you have a favorite AI tool? Do you have a go to that you really love? 

**Alicia Wright:** If we mean like specific models, I love Claude 3.5 sonnet and oh, but I've recently come to love. I guess the answer is no because , because now I have a second one.

GPT o1. ChatGPT o1. I love o1. I've started using it. More for outlining and to help me plan my chapters because I plan I do scene beats. I probably should have said that when talking about my workflow, I do scene beats and then I basically talk them out.

So I use [00:28:00] o1 to help me generate scene beats. This is a new development using o1 for that. I think I've been using o1 for all of two weeks and I'm obsessed. 

**Danica Favorite:** I love that. I think that that's a really good point though, is recognizing that the different models have the different talents.

**Steph Pajonas:** O1 is got that thinking in the background and the reasoning part. So it's really good at brainstorming. And developing an entire outline for a book that is a complex task, right? You have to consider all the moving parts in your outline before you start writing, especially if that is your process, going from outline to beats to actually writing. Okay. So that is a complex process and it is hard for a lot of the regular models to deal with that because it is a lot of thinking out and projecting out towards the future for that story.

But the reasoning models that we're getting to see where you've got GPT o1, they're working on o3 right now, and then we've got some other [00:29:00] reasoning models. We've got now DeepSeek out of China and a lot of other ones that are working on the backend with tokens and thinking out the process before it actually gives you a response.

I think that's great because I think O1 is going to be awesome for authors when it comes to all of the brainstorming and architecting your story and getting it right. And especially if you've got subplots going on, the reasoning models are going to be great for that. 

**Alicia Wright:** Yeah, I'm excited.

I'm not outlining a new project right now because I'm writing. I'm going to finish the project I am working on. But for real, but I'm excited to use it for outlining the next time when I'm outlining my next project. 

**Steph Pajonas:** It will be there for you. I'm sure it will. 

**Danica Favorite:** And I think it'll be better. I think that's what's really cool is like right now we're like, Oh, yeah, it's really good at this and can do these cool things.

But they're always getting better and it's really neat to be able to see it. Okay, this is reasoning through and thinking through, especially like where you were talking about in your process, how needing [00:30:00] that connection. The reasoning models are capable of saying, okay, I've got point A and point C, what's the logical point B?

We have said multiple times, and this is still true, the AI can't legit think. However, the reasoning models do go through some of those logic processes that mimic thinking and can say, okay, this based on the data I have this is the most logical piece here.

**Alicia Wright:** And I often like to ask AI models, why I, I need a suggestion to connect these two things that I want to do, give me such and such suggestions and tell me why each one works. And sometimes it will give me nonsense, and I know not to consider that suggestion, but if it has a good reason, then obviously some of these, the four old models say doesn't validate its work, but I feel like there's something going on if it has a good reason versus when it doesn't have a good [00:31:00] reason. And obviously I can analyze their reasons and that helps me make my decisions. 

**Danica Favorite:** That's really. I think this is my, cool tip to take away from this is asking it why it came up with that conclusion.

Like I hadn't ever thought of doing that. And now I'm like, Oh, that makes so much sense. Like, why did you do this? Or why do you think I should do this? I think, again, you're talking about the validation, like that is a great way to help you validate what that output is, because it is important, of course, to validate your output and that's going to be a cool one to take away. 

**Steph Pajonas:** Oh, yeah, I'm putting that one in the notes. Don't forget to ask why. Why? And be like a five year old. 

**Alicia Wright:** O1's answers to the why questions are much better than 4o's answers to the why questions, which is one of the things that got me super excited about it.

And now I want to use it for everything. 

**Steph Pajonas:** Oh, interesting. So it had a better response, more logical response, did you feel? 

**Alicia Wright:** Mm hmm. 

**Danica Favorite:** Yeah, I think that's even good, for people who are [00:32:00] using it for marketing, which we haven't even talked about. And I don't know if you're using it for marketing or anything like that.

But, hey, create a marketing plan. Okay. Why do you think this is a good marketing plan? I'm really excited now to just try this myself because I don't know if you use it in your day to day work, but I'm using it constantly in day to day work, and even sometimes I'm like, well, that's so weird. I wonder Just to ask it, why is like, wow, this could get me some really cool stuff.

**Alicia Wright:** Luckily, my manager at my day job is really excited about AI too. So I get to use it at work and then I show him when I did something cool. And he's like, yeah, this is so great. And, and yeah, that's, I love that. It really makes me love my job. 

**Danica Favorite:** Yeah, I have the same thing at work. I know Steph also has the same thing at her job, because, 

**Steph Pajonas:** Well, my, my whole job is AI.

**Danica Favorite:** Really? 

**Steph Pajonas:** And AI and writing. 

**Danica Favorite:** Yeah. But, and I think that's what's cool about our AI for Authors group on Facebook as well, is, is really, when you have those cool AI discoveries to be able to share them. [00:33:00] And we always want to hear what people are doing with AI. And what they've learned and what's really cool because, obviously, Steph and I work with AI all the time and here's Alicia giving us these really cool hot tips.

We're like, I didn't know that. And that makes sense. So they always be curious and learn and find those cool little nuggets of things that you can take away. 

We're getting ready to wrap up here. Alicia, is there any other little nuggets that you have or things that you want to share that maybe you didn't get a chance to cover or you think would be really important for our audience to hear?

**Alicia Wright:** I guess I'd like to talk really briefly about the fair use issue. 

**Steph Pajonas:** Oh, yeah. 

**Danica Favorite:** Oh, yes, 

**Alicia Wright:** I know that folks who are against using AI, especially in the writing industry, are really confident that using copyrighted works for training AI is not fair use. And I'm not going to say I'm I 100 percent think it is because you never know what the courts are going [00:34:00] to do. And our court system has done some unpredictable things in the past couple years.

But the case law suggests that the fair use argument is very strong for the AI companies. There has been case law where into the entirety of copyright work has been used in a way that the courts decided was transformative. And even though it was commercial, it was considered fair use. 

And in a lot of those cases one of the things that helped the user went out was that transformative factor as in what you're creating is very different than the copyrighted works that went into it. And that's what's happening with AI and I really think that that despite the fact that entire works have been used despite the fact that is the commercial use. The work is so transformative. And at the time [00:35:00] when these AI companies were first using these copyrighted works, there wasn't an existing licensing system in place, which is also a factor that's considered have you stepped on an existing method where you could have gotten everything you need.

That that's a factor that didn't exist yet. 

And the transformative use. I feel like they the case is really strong for fair use. I'm not going to say I'm 100 percent confident, but I'm somewhere around the 80 percent confident that this is going to land in fair use territory. And I'm really interested to see how the writing industry would react to that. when that happens. 

I also think that because that's my prediction, if you're being offered an opportunity to license your work for AI training, now is the time to jump all over that because if you're finding a fair use happens, some of those opportunities are going to dry up real fast. 

**Steph Pajonas:** I agree with that. If you're going to be offered a licensing deal to have [00:36:00] you give over your work to use for training a new AI model or something, this is the time to take that deal. Because I agree with, I agree with you. I'm not a lawyer, obviously. I've been watching this happen, the case law and whatnot.

And I'm also at 80%. So I will take your 80 percent as well. And I feel pretty confident that. That they had every right to use it to train. 

I also I have strong feelings about the fact that we are supposed to level up as a species and that we need to tap into all of the stuff that we've done already in order to grow and in order to evolve into whatever is next for us, in both technology and in our personal lives and whatnot. And I don't think that you can do that by just squirreling away all of our works and holding it back from society. So I'm very much towards like, here, take all of my work, use it. I want us to grow. I want us to evolve. And I'm not afraid of [00:37:00] that. So... 

**Danica Favorite:** That's why I became a writer is I want my work to help people and if it can be used to train AI to help people use language better or whatever. I am here for that. I'm all for it. I really like what you were saying about evolving as humans because we also have to evolve as writers and to me I know there are people who would call using AI for your writing lazy. I completely disagree on so many levels because what I have found is it's making me more creative as a writer.

It's making me analyze things better. And I know Steph and I did an episode a while back, so please listen to that, where we talked about how using AI has helped us grow as writers. But I really do encourage you to use this as that challenge to how can I become better, and how can I use these tools to improve what I'm doing, improve my process.

And again, if you don't want to use the AI to write, Okay, cool. Use it to [00:38:00] take some stuff off your plate so that you can devote more time to writing and getting better at your craft or whatever that is. So, that would be my little soapbox. 

But I do want to actually take it back to what you were saying, Alicia, about transformative, because I know that we're all very familiar with that term and what that means.

But a lot of writers don't really know what it means. And so, could you just explain what it means for the work to be transformative? 

**Alicia Wright:** Transformative essentially means that the what you produce transforms its inputs. So it's different. It's completely different. So we're talking about in an AI model being trained on copyrighted works, the AI model itself is so completely different from the copyrighted works. It is not directly competitive with the copyrighted works. I can't take Chaucer back out of the AI model. Well, maybe I can because it's so popular. I could ask it to [00:39:00] generate Chaucer. 

But most of the works that go into it I can't take them back out. And I can take this AI model and create so many different things. I can help the medical field. I can help in cyber security. There are a hundred different fields that this AI model can benefit. And because of that, it is the thing that is so different from the copyrighted works that went into it.

That's really what fair use is about, is about growing and being bigger than the copyrighted works that you used to create your thing. These AI models are so transformative when compared to the copyrighted works that went into them. It would shock me to hear that that is not fair use.

**Danica Favorite:** Yeah, I think that's really a good point because, we obviously know all of the books that are inspired by other books and that's considered fair use. Those, you can see those close correlations. between the different works and [00:40:00] say, okay, yeah, this is where this concept and this concept, where they intersect and have been transformed.

But like the AI takes that, I think a thousand steps further than just, hey this book that I read inspired me to write this whole series. And that part we've always said is totally fine. So like, we're just taking that 10 steps further. 

**Steph Pajonas:** I have a whole series that I wrote that was inspired by Bridgerton, but there's nothing remotely Bridgerton about it, except for the fact that it was about a bunch of sisters, So, everybody takes all of their, inspiration from other places and then transforms it into their own work. And that is what AI does as well. It's transforming. It's using a language base in order to give you something new and exciting. 

Well, thank you so much for coming today, Alicia. This is such a good talk. I think that people are really going to devour this topic and probably, come back and let us know how they felt about it. 

It would be great if people came [00:41:00] by Brave New Bookshelf. com and give us your thoughts and what you're thinking about, especially for your own writing and everything that we talked about.

Alicia, tell us about where people can find you online. 

**Alicia Wright:** I write as Alicia Ellis and I write mostly mystery and science fiction and science fiction mystery. And you can find me on social media as Writer Alicia everywhere. All social media platforms and my website is WriterAlicia. com. 

**Steph Pajonas:** Excellent. We'll be able to put that in the notes as well so that people can come and find you.

Everybody who's been listening, come, Like our Facebook page, like and subscribe on YouTube. Come click those buttons so that you can be alerted when we get new episodes out. Danica, do you have anything left that you want to say before we say goodbye?

**Danica Favorite:** Yeah, just, once again, I want to encourage all of you if you have comments, if you have thoughts, questions, things you want us to cover, please make sure [00:42:00] you're leaving us those in the comments because we really do listen. If you're listening to all of our episodes, we just did an episode that was dedicated to a listener question, and we would love more of that.

For us, this is our passion to help educate people and bring in that knowledge so that when you make whatever decision you make regarding AI, it's an informed decision with actual facts and experts . So, please. Submit your questions, share things. Our goal is education. And so if we can help you, then that is a win for us. 

I'm excited. We've got some really cool stuff planned, but I'm always eager to do more to serve the audience. 

**Steph Pajonas:** Absolutely. So come by bravenewbookshelf.com, read the show notes from this episode, leave a comment. And we're excited to hear from you. So, from everybody else here, we're gonna say goodbye, and we'll see you next week.

Okay. Bye. 

Thanks for joining us on the Brave New Bookshelf. Be sure to like and subscribe to us on [00:43:00] 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.