Join the discussion: Ask Matt and Jordan questions about AI and creative content
Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Connect with Matt Thomas: LinkedIn Profile
In today's fast-paced digital world, where automation and artificial intelligence (AI) are becoming increasingly prominent, writers and content creators face new challenges and opportunities. The recent episode of the Everyday AI podcast, titled "Writers and Content Creators: Future Role in a World of AI," delves into the implications of AI-powered writing tools and their impact on the future of the industry. In this article, we will explore the key takeaways from the episode and discuss how writers and content creators can leverage AI while maintaining their unique voice and creative abilities.
The Power of AI-Powered Writing Tools:
The podcast episode highlights the conversation between host Jordan Wilson and guest Matt Thomas, where they discuss the benefits of using chat-based AI writing models, such as Chat GPT. These tools provide a valuable resource for writers and content creators, offering assistance in various writing tasks, including ideation, word choice, and editing. They allow for increased speed and efficiency, aiding content generation and optimization.
Custom Instructions and Prompts:
One notable aspect discussed in the episode is the concept of custom instructions. By providing specific instructions to AI models, writers can customize the output to match their unique writing style and preferences. However, it is essential to strike a balance and avoid overreliance on AI-generated content. Writers should treat AI as a tool to support their creativity, rather than a replacement for human ingenuity.
The Importance of Prompt Engineering and Clear Writing:
Prompt engineering, as mentioned in the podcast, involves crafting detailed and well-structured prompts for AI models. This process aligns with the principles of good writing, emphasizing the need for specific, clear, and concise language. Matt Thomas highlights the parallel between prompt engineering and good writing, suggesting that writers who excel in their craft are well-suited to leverage AI effectively.
AI as a Writing Partner and Critical Analyst:
One intriguing discussion in the episode revolves around writers utilizing AI as a writing partner. AI models like Chat GPT offer a valuable second opinion and an objective outside perspective on crafted content. They can assist in ideation, offer alternative framings, and point out weaknesses or superfluous information. Writers can leverage these AI tools for inspiration and assistance in refining their work.
The Future of Writing and Content Creation:
As AI continues to advance, the future of writing and content creation seems ever more intertwined with generative AI systems. The podcast hosts express excitement about the creative possibilities AI offers. However, they emphasize the importance of writers and content creators having a solid foundational writing background. The ability to effectively leverage AI tools not only enhances efficiency but also ensures the production of high-quality and engaging content.
The episode of the Everyday AI podcast has shed light on the evolving landscape of writing and content creation in the era of AI. Writers and content creators must embrace AI as a valuable tool to enhance their productivity and creativity, rather than viewing it as a threat. By incorporating AI-powered writing tools and adhering to the principles of good writing, writers can amplify their skills and deliver compelling content that resonates with their target audience. As AI continues to advance, leveraging its capabilities will become an essential aspect of success in the writing and content creation industry. Stay informed about the latest AI developments by signing up for the Everyday AI newsletter and explore the abundant opportunities awaiting writers and content creators in the age of AI.
Topics Covered in This Episode
1. Leveraging AI for writing and content creation
2. Using ChatGPT for word choice and ideation
3. Advice for writers and content creators leveraging AI
4. The future of writing and content creation with AI
Jordan Wilson [00:00:17]:
What's the future of writing and creating content now that we have these generative AI systems that do it so quickly? It's honestly something I think about all the time. I've been getting paid to write for 20 years. I was a former journalist. And it's something as we see developments and generative AI and the type of writing and content they produce, it's always important to think and talk about the future. And that's exactly what we're going to be doing today on Everyday AI. So welcome. My name is Jordan. I'm your host.
Jordan Wilson [00:00:49]:
And this is your Daily Livestream podcast and free daily newsletter bringing you the absolute latest and greatest in what's going on in the world of gen AI. So this isn't just about learning and hey, here's all this stuff that's going on, but we break it down and give you practical and tactical advice on how you can actually use what we talk about to grow your company, grow your career. All right, so a little bit different. A change up today. We got to spice it up here on everyday AI. So not everyone can do a 07:30, a.m. Central Standard Time zone. That's early for a lot of people, especially if you're in different city, different state, different country.
Jordan Wilson [00:01:30]:
So this is still going to be debuting live. Don't worry. But it's actually technically pre recorded. Hey, we won't tell anyone if you don't, right? So with that, I'm excited. If you still want the AI news, don't worry. Just go to your everydayai.com sign up for the free Daily newsletter. We're going to have the Daily News, so don't worry. You didn't forget it.
About Matt and Perfect Prose Writing and Editing
Jordan Wilson [00:01:51]:
It's just in a different spot. So let's talk now about the future role in a world of AI. How are we going to write content? How are we going to create content? So I'm not on this alone. I'm very excited to have our guest for today. Let's bring him on and welcome him to the show. Matt Thomas is the owner of Perfect Pros writing and Editing. Matt, thank you for joining us.
Matt Thomas [00:02:13]:
Thanks, Jordan. How are you doing, man?
Jordan Wilson [00:02:15]:
Oh, I'm doing fantastic. So Matt, just tell everyone just real quick a little bit about what you do know, perfect Prose, writing and editing, and maybe a little bit about your background as well.
Matt Thomas [00:02:26]:
Sure. I'm the owner of perfect pros writing and editing. My background kind of all over the place, but I've been a writer for the past 13 years. I've worked as a technical writer, a copywriter, professional editor, and most recently, last year content creator, especially LinkedIn. So, yeah, obviously I've got kind of a varied background in writing, but pretty excited about the impact of, you know.
Matt's first reaction to GPT 3 and learning curve
Jordan Wilson [00:03:02]:
I'm curious about what was because remember, you know, I talked about at the top of the show. I've been in the writing space, give or take, for 20 years. I remember my first reaction to seeing the GPT technology. I think it was either late 2020, early 2021, when GPT-3 was first released through other commercial products. And when I saw it, I was like, wait, what? Do you remember the first time that you saw a generative AI system create copy? And if so, do you remember what your reaction was?
Matt Thomas [00:03:36]:
I mean, it was not long ago, I only started really using it. I signed up for Chat GPT last December. But as soon as it kind of hit the market on a bigger scale, everyone was talking about it. Right. And it was actually one of your posts that kind of got me. I think it was either last year or early this year, where you were talking about I've been a writer for 20 years, and it pains me to say this, but I think a lot of writers are going to get replaced by AI if you don't start learning to use it and work with it. And that kind of lit a fire under my ass. And then once I started playing with it, I was like, this is pretty cool.
Jordan Wilson [00:04:15]:
Yeah, we'll do it a little differently today too. Matt, let's skip to the end, and then we can go and hit rewind. But when we talk about even the future role of writers and content creators, what do you think that is? I have a lot of thoughts, but I want to open with just like, what is it going to look like in the future for writers and content creators? Let's skip to the end.
Matt Thomas [00:04:43]:
I think in the future, the most efficient, most effective writers are going to be able to leverage AI and work with it and know how to use it for different purposes. I mean, the thing is, I think that a good writer who knows how to work well with AI is just more efficient and can be more effective than a good writer who can't. And I think that a solid writing background and skills are still valuable. Right. It's like, I'm a musician too, but if you put someone with no musical training in my studio or sit them down at my drums, they're not going to be there's. Technologies I can use, like samples and loops and stuff like that, or plugins to expand what I'm doing. But I think having the skills that base to begin with definitely makes you more effective.
Jordan Wilson [00:05:39]:
Yeah, it's actually such a great point because maybe even I personally overlook that aspect of it know, being a writer by trade, how much that actually helps once you jump into any generative AI system, maybe even talk. Matt, a little bit about your learning curve. Was it easy for you to start using? I'm guessing you're using tools like Chat GPT or something like that, but I guess. What was your learning curve like? Was it maybe a little easier than average, maybe because of your writing background?
Matt Thomas [00:06:18]:
Yeah, I think so. The thing is, I think that was good about when I started using is I didn't really have any expectations or I wasn't like, I need to I just started playing with mean, which I think is probably the best approach, and then just playing with it consistently. I mean, I wasn't super impressed by the written output of GPT and I'm still not particularly impressed by it, but I mean, there's ways I'm getting better at prompting it, but just playing with it and just being like this is pretty crazy. Like how you can just frame things in different ways and how you can just add different context to shape your responses. But yeah, I remember being underwhelmed by the actual copy output, but more just like that was the least interesting thing about it to me. It was more the creative possibilities with it.
Using ChatGPT as a writer
Jordan Wilson [00:07:09]:
Oh, I love that we'll talk about that a lot, but the creative possibilities, I think within chat GPT, especially Chat GPT when you use plugins, is endless. But I want to go back to something you said there because I'm interested and I want to dig in a little bit more. You said that you were underwhelmed with the quality of the like, especially you, right? So you have a background in technical writing and we'll include this in the show notes. Matt actually puts out great content on LinkedIn. He's always like comparing and breaking down writing to very meticulous level, right? Like word order, number of words, cutting the fat, all of that. So I want to dig in a little bit deeper into the words and into the output and the quality. So maybe in your experience, when you say you're underwhelmed with the output of what you're getting out of Chat GBT, maybe talk about that a little more, maybe what's some of those things that you're like, it could be better here.
Matt Thomas [00:08:13]:
I mean, especially when I started prompt, I had no idea what I was doing, right? When I started prompting it, I was just putting in I was impressed by it at the time, but now looking back, I'm just like, okay, yeah, but it's like anything, like you said, there's a learning curve, right? I've been using it, I use it every day, especially as a content creator. It's been a godsend in being able to put out content consistently, but yeah, just how fluffy the language was, right. It doesn't write in a way that's my preferred kind of tone. But again, I don't think I've never just copied and pasted the output of anything and posted that as a post just because sometimes but where I do find it helpful is I really like using it as a second opinion or like an assistant. That's something because it's almost like having a writing partner or just like an objective person to bounce ideas off of, which is something I find helpful, especially if you're a writer like me, who can get kind of into the details and the nitty gritty and just like this word or this word, I can't decide. It's just one word, but all those details add up. So I need, like a second. Yeah, I like having that, like a second opinion, basically.
Jordan Wilson [00:09:35]:
Yeah. And I'm going to go ahead and reframe this and let's talk about it a little bit more. Because, Matt, like what you said there, I think that's where so many people, one, they either give up on Chat GPT when they see the output is not that great. So it's great to hear for other people to hear your advice that's saying, like, oh, no. Even though I was maybe not super impressed by the output originally, I still kept going at it and found maybe other or better use cases for your particular role. But one thing I want to talk about is a lot of people think of even Chat GBT as an extension of theirselves, which it absolutely can be. But one thing I like to tell people is think of Chat GBT. And you kind of reference this here, Matt, is think of it almost as not just a second opinion, but I always like to tell people, think of it as an outside entity.
Jordan Wilson [00:10:28]:
So instead of collaborating or having this be 10% of your work or 50% of your work, you be 100% and have Chat GBT be 100% and treat it as a completely separate entity. So maybe talk a little bit more about that approach. I'd love to hear.
Matt Thomas [00:10:45]:
No, that's actually probably the best way to describe it, because just conceptually, I've never really looked at it as a replacement for writing for my writing. So I think that is actually probably the best approach, is like, I'm still going to write it, I'm still going to put it 100% of it in. And that's like, one thing I think as well, is when you come at it with a strong idea, like a strong initial idea, I find, and then use it to kind of refine that, I find it's more effective. And that's the best way to put it is that I think it's least effective, or you're not really using it to its full potential, or well, if you just see it as a replacement for having to write. But like you said, if you're coming at it 100%, you're going to put the effort in. It's a great way to kind of amplify that and to enhance it.
Jordan Wilson [00:11:35]:
Yeah, absolutely. Again, I'm going to keep digging down here because we actually have some great resources and other shows that go along with this. But I am here curious, and this is obviously, if you're listening, this is two very wordy people, right? Like two people that word nerd writing. Yeah, love the words, but maybe even with some of your disappointment in the output, is it just kind of like, hey, does it sound too robotic? It sounds like too run of the mill. Or maybe it just doesn't sound enough like you. Or maybe, what is that thing, at least right now with where you're at, that you're like the output, at least for me, being a writer, technical writing background, that it just doesn't stack up.
Matt Thomas [00:12:17]:
A lot of times it's just too fluffy. You know what I mean? One of the things I've done recently that's helped a bit with that is adding custom instructions to GPT four. So kind of giving it some constraints. Like, here's my background. I'm a LinkedIn content creator. I'm also a writer, an editor. Don't limit things like it loves rhetorical questions, doesn't it? So a lot of those don't ask rhetorical questions, keep it at around a 7th grade reading level. Don't use more than, I don't know, twelve to 15 words, absolute max in a sentence, that sort of thing.
Matt Thomas [00:13:01]:
So just trying to you're right. To make it sound more like me, I guess when I read it, or when I read the output a lot of times, that is kind of what I find, is that it just doesn't sound like my tone of voice or the way I would write.
Jordan Wilson [00:13:15]:
And for those of you joining us here on the live stream, so Matt's talking about custom instructions, which I think there's great pros, but there's also some essentially, you know, like he was saying. So instead of having to do a little bit of training each time you start a new chat and chat GBT, you can essentially tell, hey, so the two different boxes where you can say, what would you like chat GPT to know about you to provide better responses? And you can write about 1500 characters there. And then also, how would you like chat GPT to respond? The same thing you can put in about 1500 characters. So is this something that, Matt, you at least in your use cases, that you're finding custom instructions helpful?
Matt Thomas [00:13:59]:
Yeah, I think so. I think it's just most helpful to provide some background, some context. I mean, I can see, though, if you have to write in a bunch of different tones of voice. Like, I have a friend who's she's she's an SEO blog writer, and she frequently has to use different tones of voice. And that's one of the reasons she doesn't kind of like the custom instructions, because she's not just writing one way, whereas I tend to write in one style. But, yeah, definitely. That is one potential downside, I think, if you have to vary your tone a lot. But, I mean, I find that regardless of what I'm writing, I tend to write kind of the same way, for better or worse.
Matt Thomas [00:14:37]:
For me, it is helpful and just like, yeah, context. Even the other day I prompted it or I was just asking it something and then it was just like, oh, given your background, I was like, how do you know I'm a LinkedIn? And it was like custom. I was like, oh yeah, I forgot that. I didn't realize it just was just a reminder. It's referencing that context in any of your interactions with it.
Prompt engineering enhances responses
Jordan Wilson [00:14:59]:
Yeah, it is. And I think Matt, in your use case, it's perfect for that, right? Like custom instructions. Personally, I'm not a fan because I use Chat GPT for just about literally anything and everything you can think of. So sometimes those custom instructions, if you use Chat GBT all over the place, it can actually be like, wait, why am I getting responses all like this that are not really in line with the prompt? And that can kind of be the reason why. But speaking of prompting, Matt, kind of this term that's floated all over the place now is prompt engineering. And I make fun of the term a little bit because is it a real thing? Yes, but all prompt engineering is to me is knowing how to properly work with a large language model and getting the most out of it. But maybe even with your background as a writer, did you find prompting easier? Like, were you all of a sudden like, wait, my writing background? Hey, maybe AI is going to take some writing jobs, but I can really get in and use Chat GBT and these large language models at a whole nother level just because of my background in writing.
Matt Thomas [00:16:10]:
Yeah, and that's actually something I didn't really that's actually just something that's kind of occurred to me lately, is that as I've been digging into more kind of prompting, I'm taking some course. I'm taking Rob Lennon's AI content reactor course right now. I signed up for your prompt polish course as it like, as I was reading some of this stuff in the AI content Reactor course, I was well these he was talking about the characteristics or the principles of good prompt writing. I was like, well, these are all just principles of writing. Well, so like, using specific, clear, concise language, using language, and kind of orienting your prompt, aiming it at a specific purpose or goal, I find same with content. Some of my best posts are ones that start with like a strong initial idea and I just run with it. Same thing with prompts. I find my typical use case is to come at it with an idea already, not to just kind of try and not to use it to try and think of stuff.
Matt Thomas [00:17:13]:
But I find that that's more effective. Like when you come at it with a strong idea and then have it expand on it, it already has kind of a direction to go and you're giving it kind of some constraints and shaping that response. Also something else I like to do in my content, but like using examples or analogies to illustrate my points clearly. Same thing with prompts and just really minimizing. You got to think of it as stupid, right? You got to think it doesn't know anything. You want to be as explicit and as clear and minimize ambiguity as much as possible. I guess. For me personally, one of the advantages that my technical writing background gives me is I'm really used to writing in a structured, clearly organized way.
Matt Thomas [00:17:56]:
Like, pretty much everything I write, even if it's just like an email or something quick I'll write, even if it's just handwritten a quick outline, just because I like to have it organized. And I found that Chad GPT really loves outlines, especially when they're kind of, like, structured and you break stuff down in a well organized, hierarchical so, yeah, I think that is one inherent advantage that writers have when it comes to prompting any large language model.
Jordan Wilson [00:18:23]:
Like you said, it's funny, Matt, because you said something there. You're like, oh, prompting.
Matt Thomas [00:18:29]:
I just read this book recently.
Jordan Wilson [00:18:30]:
Yeah, prompting is essentially writing well. It's actually one of my favorite books on writing, so I'm sharing it. If you're joining us on the podcast, throwing it up on the screen here.
Matt Thomas [00:18:39]:
Excellent book. Highly recommend.
Jordan Wilson [00:18:41]:
It's called On Writing Well by William Zinzer. And it's funny because this book was originally from the they updated every decade or so, but you could just put some duct tape on this and say, Prompt Engineering 101, right?
Matt Thomas [00:19:00]:
Yeah, you totally could, actually, because that's exactly what he talks about, those principles of writing well, right? Like, clear, use precise language, minimize ambiguity, use examples and analogies. Great book.
Future of writers and creators with generative AI
Jordan Wilson [00:19:18]:
I'm also curious, Matt, even for you personally, right? Because I have an agency right now, we're mainly focusing on generative AI, obviously. But even with your background in technical writing and now this newfound understanding of, hey, I'm a great writer, I can translate that to generative AI and probably get way better outputs than the average person, how has that even changed or shaped your outlook on your career.
Matt Thomas [00:19:55]:
Jordan Wilson [00:19:56]:
Has yeah, not yet.
Matt Thomas [00:19:58]:
No, it's definitely, like, especially lately, AI is the future, right? For better or worse, it's coming. But I'm also excited about it, just, like, the creative possibilities. It's just interesting, it's just cool, and it also just feels like a space where there's a lot of it feels to me like I could come up with something new or innovative because it's, like, wide open, right? It's so new that nobody's I mean, there's experts in it, but what they're an expert in is constantly changing, and the entire world of it is constantly changing. So it's definitely like I mean, I'm really glad that I kind of made the jump on it when I did, and I'm telling everybody, I know this is the future, man. And it's just efficient if you're creative and kind of lazy like I am and you want to do things in the most efficient way. AI is awesome.
Jordan Wilson [00:20:51]:
Yeah. Especially when you can refine those outputs and get them exactly how you need them or how you might want them. Matt, you said something there and it got me thinking back to my kind of initial learning curve, right? So GPT-3 kind of the base technology was rolled out in a bunch of other third party softwares before chat GPT, so some early ones were like copy AI writer Jarvis, now called Jasper. And at least for me, I had this weird feeling where when I saw the output and I worked with it a little bit, my background as a writer almost felt like threatened, like, wait, what? This is pretty good. Have you ever felt that way yet? And if so, maybe how did you respond to it?
Matt Thomas [00:22:19]:
No, not honestly, maybe I should. Like I said, you scared me last year with that post, but that was like a good thing anyway. But no, I'm not really threatened by it. But the thing is too, I think that's important to remember is that people like you and me, we're using it every day and we're in this world, but there's so many people who haven't. Even the fact that I use it so much, I don't consider myself an expert by any means. But compared to probably like a large percentage of the population right now, or even people my age, I relatively am an expert, even though I don't think of myself as one. No, I didn't feel threatened. Maybe that's my own hubris.
Matt Thomas [00:23:01]:
But I did think I was like, this is going to replace a lot of mediocre writers. That was kind of my first thought. A lot of kind of mediocre content writers and blog writers and stuff. That's why I think you need to have other skills. And I guess a good thing is that my background, I didn't just do one thing, I kind of have copywriting and technical writing. So a bit more of a holistic approach to it, I guess. I'm not just doing one thing that can be replaced.
Matt's experiments with AI vs human
Jordan Wilson [00:23:29]:
Yeah. And it is worth mentioning. And again, Matt's content is great. We're going to have a link in the show notes to follow. I know you're like same thing with me, you go through ebbs and flows, but he has so many great PDFs carousels of just the very highly technical nature of writing in general. But I think you also did I'm trying to remember Matt you did like the is this which one's better at editing, like a human or AI.
Matt Thomas [00:23:58]:
Right, yeah, that's was what was kind.
Jordan Wilson [00:24:02]:
Of the general takeaway from those early experiments because, yeah, you kind of broke down long kind of extra wordy sentences or paragraphs, and you went through, I think, and cut them down. And then you also use, like, AI. Right?
Matt Thomas [00:24:18]:
Yeah. So I used a software called Unfluffer. So I had basically a big, long wordy phrase and then showed the original and then two edits, one me, one versus a computer, but didn't tell you who was who's. Kind of had people go through them in the carousel, decide which one they like better, and then revealed who is who. So man for AI one, that was the final score.
Jordan Wilson [00:24:42]:
Man for AI one. All right, so AI. Didn't get blanked out, but human crushed it.
Matt Thomas [00:24:48]:
Yeah, human prevailed for now.
Jordan Wilson [00:24:51]:
We're still there. Hey, humans, we're still there. Although, actually, speaking of that, I've actually wanted to do this where I partner up. Who knows? Maybe I'll do it with you or someone else. I partner up with a great writer, a great creator. I use AI to write posts as them. And then the human writes, and we put it out there, and we have maybe we'll do that. Maybe we'll do that and not tell anyone, okay, let's get away from writing because not everyone out there is reading William Zenzer book twice a year.
Matt Thomas [00:25:24]:
Not everybody's nerds like us.
Creative AI use cases
Jordan Wilson [00:25:26]:
No, not everything. So let's talk, like, creativity a little bit. So, obviously, I think one of the things people always think about chat GBT is, oh, to help me write better content. But a lot of people either aren't very good at it, or they're not good at getting better outputs. But what other use cases specifically have you found that are super helpful for you? Maybe some creative use cases that aren't just like, hey, write me a blog post on this. What are you finding real value in as a writer specifically for these creative use cases?
Matt Thomas [00:25:57]:
Well, it's funny because that's actually the writing is like I don't think I've ever just really used it to write me a post, but my primary use case for it is I really like using it as, like, a second opinion, like somebody to bounce ideas off of, like a writing partner. So a lot of times, especially if you're somebody like me and you kind of are really in the nitty gritty about the details of writing, sometimes it'll come down to the same sentence framed two ways, but two words are different. Or is this word better? Is this word better? I like to use it as an interactive, like a virtual thesaurus that you can interact with. So which one of these words fits this context better? Because sometimes it's hard to decide, right? And it's nice to have an objective outside opinion, especially when you get too attached or you've been spending too long on something. Yeah, like I said, the second opinion like an assistance expanding on an outline. Like I said, a lot of times, I start with an So and Chat GPT loves really structured outlines. So I'll type up an outline and I'll feed it into GPT and I'll break down the sections and all the sub points and stuff like that. And I find with a really structured by structuring it really clearly like that, it tends to provide better outputs in terms of ideation and talking points, finding different ways to frame an idea.
Matt Thomas [00:27:26]:
So if I have an idea for a post about this, how are some ways I can frame it in a way, but make it relevant to my audience in a way they'll care about? Even last week for my newsletter, I put out it's kind of exploring the parallels, like, I'm a musician as well between drumming and writing. So that was kind of something like I had the initial ideas, but some of those talking points, like the sub points with Chat GPT, I fed that outline in and had it expand on the ideas a bit.
Jordan Wilson [00:27:57]:
Yeah, it's funny because this goes back to what we were kind of talking about earlier, is thinking of Chat GBT almost as an outside entity and as a second person and not as an extension of yourself. And Matt, you mentioned the prime prompt polish the PPP course, and I was actually shocked early on when I was really applying even my writing background to prompting and kind of like, quote unquote prompt engineering. If you tell Chat GBT to almost work outside of you and to pick apart your ideas and your strategies and your writing, there's things like you just said you're a professional musician and a professional writer, and it was creating parallels that maybe you weren't even thinking of, right?
Matt Thomas [00:28:43]:
Yeah, and actually, that's another one, too, is like analyzing my writing, what's good about it, what's bad about it, especially what's bad about it? Or if I'm like, Is this too long? What could I cut here? It helpful to kill your darlings sometimes, right? It's like an objective outside hatchet man to come trim all the fluff away.
Jordan Wilson [00:29:07]:
Oh, I love it. But I'm also thinking of Matt, just like a fluffy giant robot with an know coming to chop, cutting, cutting all the fat from our wordy prose. Okay, so, Matt, we've covered literally so much. We've talked about what you think the future is going to be for writers and content creators with AI. We've talked about how being a great writer to begin with is going to lead to better prompts, and even talked about a couple creative use cases using Chat GBT outside of writing. But maybe for someone else who's listening to this, who's kind of in your shoes, they're like, hey, I'm a great writer, but I haven't maybe gone all in yet. Or I'm. Only dipping my toe in this whole generative AI thing.
Matt's final takeaway
Jordan Wilson [00:29:54]:
What maybe piece of advice might you have for them? Or maybe a takeaway that you want people to leave this show with as they try to better leverage generative AI?
Matt Thomas [00:30:05]:
Yeah, sure. I think a couple of things I would say just start just start playing with it and don't have any expectations because I find that's the best way to learn is just experiment with it. Don't just rely on it for the written. I mean, I think if you're a good writer, you're going to have certain standards anyway for your writing. But play with it. Yeah, hopefully play with it, explore it and just figure out ways to use it that aren't obvious necessarily because there's tons of useful ways. And I think the other thing is, too, that we're so overwhelmed. Like there's so many AI tools all the time coming out every day.
Matt Thomas [00:30:49]:
I think just pick one or two and just start learning them. You don't have to learn every tool. Pick one or two and develop the basics. I pretty much just use Chat GPT Mid Journey on Fluff for a couple of others. I know I'm not like you. I remember when we had met, you were just like, you have a subscription to like 20 of them.
Jordan Wilson [00:31:09]:
Yeah, too many, too many.
Matt Thomas [00:31:10]:
But now I think that I've used it enough that I think I understand the basics of it, that if I pick up another one, obviously there's going to be variations and maybe a different but I think I know how to prompt a large language model now, right? Like those basics. And actually one more takeaway, actionable takeaway. And this actually came directly from you, Jordan, from when we met that time. And I don't know why I wasn't doing this before, was start a prompt library. Like a after. I was like, why am I not doing that? As soon as we had our call, I started one. Because that's what you need is like when you come up with something good, put it in there, everyone's giving away. And actually, even for my newsletter, that's what I did.
Matt Thomas [00:31:50]:
I gave away my prompt library with eight of my best kind of custom prompts that I developed for content. Because everyone's giving away prompts, but what people need is one central place to put them all and refer back to them. So, yeah, those things play with it, learn the basics, pick one and just sort of start small and focus on it and get good with that. And then try and apply those principles to other AI tools and start a damn prompt library.
Jordan Wilson [00:32:18]:
Get it going. My gosh, so much good information there. Matt I love how we wrapped it all up. Just very actionable steps on how people, whether they're great writers or not, I think we can apply those steps to just about anyone. Matt this was a great conversation I had a great time. Thank you so much for joining everyday. AI live.
Matt Thomas [00:32:39]:
Yeah, it was great. Thanks, Jordan. All right.
Jordan Wilson [00:32:40]:
And hey, as a reminder, as a reminder, yeah, you can still get your AI news. We didn't share it here live today, but make sure you go to your everydayai.com. Sign up for that free daily newsletter. We're going to have the AI news, everything else that's going on, and we're going to break down more specifics on exactly what Matt was talking about. If you couldn't write fast enough, all the notes, don't worry, we got you. Thank you for joining us and we hope to see you back for another edition of Everyday AI. Thanks, y'all.