As artificial intelligence becomes ubiquitous, artists and writers are creating their own guiding principles for where and how they use the technology in their creative practices and their work.
Tracy Hoagland is the chair of the Portland chapter of Willamette Writers, and by day, she works in the tech industry with AI. She says she doesn’t use AI in her writing practice, but has to consciously separate the two areas of her life. Jessie Kwak is an author and ghost writer based in Portland, who uses AI to supplement her writing work. They both join us to discuss their own ethos around AI in the world of creative writing.
Note: The following transcript was transcribed digitally and validated for accuracy, readability and formatting by an OPB volunteer.
Dave Miller: From the Gert Boyle Studio at OPB, this is Think Out Loud. I’m Dave Miller. As AI becomes ubiquitous, artists and writers are creating their own guiding principles for where and how they use the technology in their work. For many people, it’s not as simple as a complete embrace or a total prohibition.
We’re going to get two perspectives right now from two writers. By day, Tracy Hoagland works in advertising and marketing. She also writes thrillers and is the chair of the Portland chapter of Willamette Writers. Jessie Kwak is a ghostwriter, editor and book coach, who has also written books of her own. It’s great to have both of you on Think Out Loud.
Jessie Kwak: Yeah, thanks for having us.
Tracy Hoagland: Thank you.
Miller: Tracy, first, can you describe your day job?
Hoagland: Sure. I work for a marketing technology company. What that means essentially is we are helping businesses find their customers and advertise to them. That’s what my specific job is, advertising operations.
Miller: In that context, how much is AI used?
Hoagland: It is mandatory. We have AI principles. It’s a huge part of our job. If you’re not using it in almost every aspect of your job, you are falling behind.
Miller: OK, so basically everything that you put out to help a business communicate with their customers or other businesses is AI-enabled. AI touches every piece of it somehow.
Hoagland: Enabler is a good word.
Miller: I almost didn’t want to say it because it seems just like jargonny. But what are examples of how you’re using AI in the course of your day job?
Hoagland: Streamlining workflows is a huge thing. So looking at our processes and [asking], what could we be doing better? The big focus on it is improving efficiency, so it’s just, what can we do to make things go live faster? What can we do to cut roadblocks? Things like that.
Miller: What about generating prose that’s going to end up as the final copy for something?
Hoagland: Yeah, there’s marketing content that they use. I’m not on the exact marketing team. But yeah, I know it’s something that’s encouraged, using it to generate content and things like that.
Miller: OK, so that’s your day job. And then the rest of your life, you’re a writer in your own right and the chair of this community, the Portland [chapter] of Willamette Writers. So to what extent do you use AI in that part of your life?
Hoagland: I don’t. I don’t at all. I turn that side of my brain off. Yeah, for me, AI is like what I said. It’s all about output, improving efficiency. When I come to my own writing, it’s about the idea of lingering. It’s the idea of input. It’s taking that time to focus on the thing that you love and you care about. And you want to take that time. You want it to kind of hurt a little bit.
Miller: Jessie, can you give us a sense for the different kinds of writing that you do?
Kwak: I have been a freelance copywriter for the past 12 or 13 years. And in that, I also ghostwrite business books for coaches and consultants, and then I coach those same kinds of coaches, consultants and entrepreneurs in writing their own books. So that’s kind of the freelance nonfiction side of my business.
Miller: And being paid by others to help them write stuff, anywhere from short stuff to a whole book. And then what about your own writing?
Kwak: Yeah, so for my own writing, I write novels, I write science fiction and thrillers, and then I also write nonfiction books on creativity and productivity.
Miller: So, as we heard just now from Tracy, there’s a real division in the extent to which AI is a part of her work – deeply a part of her day job at a techie business company and almost not at all a part of her outside, personal writing life. Do you have a similarly bright line?
Kwak: No, I use it on both sides of the business. But I use it in a similar way that Tracy was talking about for those efficiency gains and things that I can streamline because I am running a writing business, and not just writing a book and hoping that it does something out in the world. There’s a lot of things like marketing copy that I do need to write that I will use AI to do generative writing for. And I do a lot of dictation, and I interview clients. So I have a lot of transcripts I’m working with, and I use a lot of AI tools for that.
But I do have a line around my creative work because, for me, like Tracy was saying, part of the point of writing is wrestling with these ideas and coming up with your own unique perspective on it. And I don’t want to outsource that to the robots. That’s the part that I love. So if there are things that I can outsource that give me more time for that work, then that’s, I guess, the line that I draw.
Miller: I want to hear more from both of you about the love of craft, even, [as] Tracy [was] saying, if it’s painful, or maybe partly because it’s painful. But to stick with the business side for a second, Jessie, how do you talk about this with your clients? I’m curious, what kind of rules – if that’s the word for it – you’ve set up for the people that you’re working for, if they want to use AI.
Kwak: I have had a lot of good conversations with clients more and more recently, as more people are getting into it. And the majority of my clients are entrepreneurs and business owners. So they’re fairly open to the idea of using AI in their business. I did have one client recently … and this is what inspired me that I was like, OK, I need to really have a policy. She came to me and she said, “OK, so I took the transcript of our last conversation and a friend taught me how to use ChatGPT to take this transcript and write the chapter.” And she said, “So I did that and it was great. It looked really professional. It was way better than I can write. But it didn’t sound like me. So I threw it away. And here’s the draft that I wrote.”
I was like, oh, I’m so glad that you actually came to that realization on your own because your readers don’t want the AI version of what it is in your head. They want your stories and your voice and your words, even if they’re not “the perfect words.” So that is a conversation I have with my clients. We’re here to get your stories out. So, let’s do that. And if there’s ways that AI can help you do that, great. But the final product really needs to be in your voice.
Miller: You know what’s fascinating about that is that that’s not a kind of moral decision matrix about the love of craft or the beauty of human striving. What your client said is that this product wasn’t quite right. This wasn’t my voice, and I don’t want to put this out in the world because it doesn’t sound like me.
I’m wondering what happens when the product is actually good enough. I mean, I think I’ve given up in underestimating the power of AI just in the last year. Every week, it gets so much better in various ways that I think it’s worth imagining the world, maybe soon, when AI products can seem soulful, can seem wonderfully, weirdly, imperfectly human, can seem like something we might actually be proud of having written.
At that point, do you think your client would have approached this differently?
Kwak: I think then it’s a matter of, are the inputs yours? Are the inputs your stories and your voice? And the thing that I do teach my book coaching clients to do is to talk out their books, because many of them aren’t necessarily writers. So I teach them, “OK, let’s talk through this chapter.” And then, if AI could take that transcript and turn it into something that they would have written that has those unique stories and their unique voice, then I would say it is their unique voice and stories. But right now, it can’t. It still wants to blandify everything.
Miller: Tracy, how do you think about that question … I guess the question is the extent to which writing is an end in and of itself? And the human act of writing and wrestling with words, and tearing your hair out, and eventually coming up with something, there’s value in that, as opposed to really focusing on the end product.
Hoagland: Well, I think that’s the question: Why are you writing? Just at the core of it. Why are you writing? Are you writing as your day job? Most writers are not marketers. Most writers don’t have the time, or the energy, or even the desire to be having to post however many Instagram posts every day, all that kind of content, because that is…
Miller: Cranking out widgets, some kind of wordy product.
Hoagland: Yeah. It’s a huge amount of content. At my company, the focus on it is we’re personalizing to an audience, right? So an audience is our customers, and we want as many customers as possible in writing. And Jessie and I were just talking about this. It’s like, you’re at the core. The story that you’re writing is the story that you need to write. And if no one’s going to like it, that’s fine, especially if it’s not your business. If you’re just writing it because you need to write this story, that’s your story, but if you’re writing it because you need an audience, AI can help with that.
AI is a prediction model. It is trying to predict how to get the biggest audience, whatever you tell it, whatever inputs that you give, but it can only look back. AI, it’s being trained on us, it’s being trained on the stuff that we put in.
Miller: The totality of human creation up to that point that it’s been fed?
Hoagland: Yeah, exactly. So I think that’s a big thing I think about, where it’s like, it can get up to this point. And I’m doing hand motions on the radio, but it can get up to this point. It’s looking back. It’s always the past. It’s always what it’s been trained on. Even at work, our engineers are using it so heavily. And some of them legitimately love coding. Coding is their writing. They love to code. And they’re using AI to make these huge amounts of code that they’re writing a little bit easier, to make that come faster.
Miller: Do you hear the same thing from some of them, that they miss the coding time?
Hoagland: Yeah.
Miller: But they’re forced because they actually can be much faster. There’s almost nothing that AI is better at, at this point, than coding.
Hoagland: Yeah, but they’ll still say that even the code that’s coming out – and I’m not a coder – it’s generic. It’s good enough. It’s that kind of code. So it’s not that incredibly efficient, actually. But the big focus of our engineers is what processes are you guys doing that are stupid and repetitive, and you want to get rid of? And that’s what they’re trying to fix, for us in particular, at the company.
But again, it always comes down to this idea of output and getting as much output, getting the efficiency in place. But even at our company, there are aspects where we don’t ever use AI. If it’s something that you require deep empathy, like a problem that requires … so, HR, right? Should HR be using AI tools?
Miller: They are, but you’re saying should they?
Hoagland: Yeah, should they? Data privacy is also obviously huge, so taking that into account. There’s just always nuance.
Miller: How much does this come up, the kind of conversations that we’re having now, or debates you hear about, from members of Willamette Writers, this community of Oregon writers? How much is AI a topic right now?
Hoagland: It’s a big topic. And like what Jessie was saying, I think there’s this gray area of, are you trying to write as your day job? Are you an indie author that is needing output? You use AI to offload those areas that you’re not the expert in.
But then it’s even things like query letters, the letter that you send to an agent to get them to hopefully read your book. Across the board, I can tell you, every writer hates writing those. They’re hard to write. You’re just condensing your 100,000-word book into 500 words or something. And people hate writing them. But we all did it, and it made you actually think about your book a little bit more. And being in that sucks, but I’m going to just keep trying, and I’m going to keep talking to my peers and getting feedback. Because that human piece, I think, is the part that you just miss so much when you try to just get AI to write it. And it’s good enough, and maybe an agent will read that query letter and it’ll be good enough.
Miller: Or is the agent going to have their AI read that letter as well. We’re having robots reading stuff written by robots, with humans in the back end every now and then giving them prompts.
Both of you, like me, are old enough that you became readers and then writers before AI was a thing. I’m curious how you think about young people now, kids now, who are growing up in a very different world when these tools, if not the norm, could become the norm for them, depending on the lessons they’re given, the prompts they’re given? The tools are readily available. I’m curious, Jessie, how you think that could shape the future of writing.
Kwak: I think, to Tracy’s point earlier about writing kind of being a struggle and being your way of struggling with ideas and what you even think, I very much worry about asking an AI what do I think? Instead of trying to understand that for yourself through the struggle of, I would say writing, but also, I mean, that’s not everyone’s way of discovering their ideas. Some people talk with friends, some people start a podcast. Everybody has a way of figuring out what it is that they think and believe.
For writers, we turn to the page, and we grab our pens and notebooks sometimes. I worry that if it just becomes standard to say, “here’s a topic that I need to understand, hey, Claude, how do I understand this,” you lose that skill of figuring out, how do you understand it? What do you believe? What is your unique point of view in the world? I do worry about that.
Miller: Jessie Kwak and Tracy Hoagland, thanks very much.
Kwak: Thank you.
Hoagland: Thank you.
Miller: Jessie Kwak is an author and ghostwriter based in Portland. Tracy Hoagland is a writer and chair of the Portland chapter of Willamette Writers.
“Think Out Loud®” broadcasts live at noon every day and rebroadcasts at 8 p.m.
If you’d like to comment on any of the topics in this show or suggest a topic of your own, please get in touch with us on Facebook, send an email to thinkoutloud@opb.org, or you can leave a voicemail for us at 503-293-1983.
