18: Using AI for Podcast Production Automation with Edward Brawer

In this episode, Chris Daigle interviews Edward Brawer, CEO of PodcastAI, a company that automates the entire podcasting process, from post-production to distribution and promotion. Edward shares insights into how PodcastAI simplifies the podcasting experience, allowing users to effortlessly upload their recordings and leverage AI to generate descriptions, chapters, key points, and even ad reads in their own voice. He also discusses the future of the platform, including plans to automate social media promotion, YouTube uploads, and even a consumer podcast app inspired by the holodeck from Star Trek. Whether you're a seasoned podcaster or just starting out, this episode provides valuable insights into how generative AI is revolutionizing the world of podcasting.

Key Takeaways:

✅ PodcastAI automates the entire podcasting process, from post-production to distribution and promotion, allowing users to upload their recordings and leverage AI to generate descriptions, chapters, key points, and even ad reads in their own voice.
✅ Edward explains how ChatGPT's breakthrough was not a technological revolution but rather the realization of how to effectively prompt and utilize existing language models, enabling more advanced AI applications.
✅ PodcastAI is already working on automating social media promotion, YouTube uploads, and even a consumer podcast app inspired by the holodeck from Star Trek, constantly pushing the boundaries of what's possible with generative AI in the podcasting space.
✅ Generative AI tools like ChatGPT and Copilot can significantly streamline coding and prototyping processes, allowing developers to quickly iterate and test ideas before implementing them in code.
✅ The advent of generative AI presents opportunities for individuals and businesses to leverage these tools for content creation, thought leadership, and even new service-based businesses, such as podcast production.

Chapters:

00:00 Introduction to Edward Brower and PodcastAI
01:16 How PodcastAI Automates the Podcasting Process
03:30 Getting Started with PodcastAI
06:28 Automated Website Creation and Promotion
09:02 Edward's Background in Podcasting and Technology
11:27 Early Exposure to Generative AI
14:09 The ChatGPT Breakthrough
17:02 Pitching PodcastAI to Sequoia Capital's  Roelof Botha
20:13 Going Through Accelerator Programs
22:08 Future Plans - Blog to Podcast, Social Integration
25:20 Using Generative AI for Coding and Prototyping
29:01 Opportunities with Generative AI for Content Creation

Find out more about PodcastAI by visiting 🔗 https://podcastai.com/


Be sure to download this week's Smartcut/s:
🔸 Shownotes & Summaries with Podcastle — https://smartcuts.io/18

And check out the related prompts at: https://PromptVault.studio

Transcript

Copy Transcript Copied

SPEAKER_00 0:52

Welcome to today's episode. We have a fascinating guest, Edward Brown, CEO of Podcastai.com. Edward brings over three decades of tech expertise and a passion for revolutionizing podcast production. Today, we'll share how podcastai.com simplifies the podcasting process from seamless post-production to automated distribution and promotion. So whether you're a seasoned podcaster or you're just starting out, Edward's Insights will help you leverage AI to enhance your podcasting game. So, tune in, get inspired, and let's get started. Welcome to Using AI at Work. I'm your host, Chris Daigle. Each week we'll be learning how today's business owners, entrepreneurs, and ambitious professionals are getting more done with smart use of tomorrow's tech. Let's get started. All right, we've got Edward Romer on the uh podcast today. Welcome. And Edward, why don't you tell everybody a little bit about yourself, but also about your cool company?

SPEAKER_01 1:57

Hey, Chris. Uh pleasure to be here. So uh yeah, I'm the CEO of Podcast AI, and we automate podcasting. So we right now we automate everything that you need in terms of post-production, uh distribution, and promotion for your podcast.

SPEAKER_00 2:13

So as I shared with you before we started recording, on Monday, I spent we have a we've we've kind of hobbled together our own version of it through automations and zap year and APIs and very complicated stuff. And it took us about 90 days. And um to explain it to the audience, it took me about 45 minutes. Now, how long does it take for somebody to do almost the same thing we're doing with your tool?

SPEAKER_01 2:40

Well, we can explain podcast AI in 45 seconds and we can set it up in less than 90 minutes.

SPEAKER_00 2:46

That's amazing. So this is the promise of AI, generative AI, right? It's like, oh, it's supposed to make things easier and faster and all that kind of stuff. And and yes, we've done it internally, but what you're doing here is off the shelf. Somebody can uh so where does it start? We talk about podcasting is a popular topic on using AI at work because AI contributes to it a lot, right? So for somebody who uh let's start with somebody who's maybe hasn't started one yet because of the concerns about technological impediments or the lift required to get all these things done. Can you kind of walk us through what that process looks like for a podcaster using podcasai.com?

SPEAKER_01 3:29

Yeah, you just open uh an account on podcastai.com, uh pick a plan, uh, create a show, create yourself as an AI host. That basically means you're creating a profile of yourself. Uh, you're uploading a three-minute recording of your voice, uh, and it now knows your voice. So it can actually produce audio in your voice, which is pretty amazing. Um, and then you assign that host to your show, you set the metadata for your show, you give a description, a title, um, you set some cover art for your show. So you just draw something in Canva or Photoshop and set that as the show cover. And now you're ready to upload your first episode. And your first episode is really just, you know, uh record a Zoom call, record in quick time a monologue, maybe in production. Uh do a Google Meet, record that. Um, tons of ways to actually record your first podcast. You don't need to start, you don't need the production quality to be super high. Um, Joe Rogan is probably the top podcaster. Well, guess what? If you look at his first hundred episodes, the production quality is not super, super high. Yeah. Right. So people will forgive you if you start without the highest production quality. So you just really just do a zoom recording, you and your friends, you and a lead, you and and somebody somebody interesting in your market. Um just record that. And we have something called an episode studio. You can upload your episode. You can, if you want to create an intro and outro, you can do that. That's totally optional. Um, the episode studio will affix that to the to the recording on both sides. It'll automatically find something that's a good cold open and sort of put that in front of your intro. Uh, and then it'll give you a fully rendered video. But that's an optional step. You can really just go go straight with your recording, upload your recording or your finished episode to podcast AI. Um, that's your first episode. You hit generate, it'll do your description, it'll do your chapters, your key points, your title, it'll do all of that stuff. We call that post-production. That's all the information on your feed that goes out to Apple and Spotify. Um, at that point, Podcast AI will generate an entire website for your podcast with all the transcripts and everything just completely automatically. Uh, for example, one of our uh big customers, uh pretty big tech podcast called This Week in Startups, This Week in Startups by Jason Calicanas. Well, this Thisweekin Startups.com that's powered by Podcast AI. That entire website is synthetically automatically created as a byproduct of using podcast AI. Very um yeah, Jason used to, you know, custom website they had to maintain and it had to be developers and all this stuff. They eventually got rid of it. Um, they just went down to a link tree, and then when I showed them, hey, we can generate this entire website for you. For them, that was a major unlock, right? They have 1900 plus episodes. We we exposed all the transcripts, everything fully searchable, just amazing for them. Um we automatically detect who are the sponsors in the in the in the in the episodes and show the logos. It's it's fantastic. They love they love the product just for that. But you know, so we just talked about doing a website automatically. We talked about the description and the chapters and all that stuff. We also do promotion. So um we make it so you can generate the top 10 viral moments for that episode. That's the top 10 moments that are the most interesting for the show. Yeah. Then you can generate the social media copy. So a beautiful LinkedIn post, uh, a Twitter thread or X thread, Instagram post, um, YouTube Shorts, all the copy that would go for there. And it'll generate the clip as well. And now you can put those to those social media platforms, and we're working on completely automating that so that you can link your accounts and schedule that out automatically. So it's really gonna be total, total automation. Um, and we're even working on uh pushing to YouTube, which is the thing all the podcasting platforms are missing.

SPEAKER_00 7:27

Yeah. Are are you uh do you have a background in podcasting or technology?

SPEAKER_01 7:33

Uh definitely technology. I've been doing websites, software, apps for the past 20 plus years, or I would say at this point, 30 years. Um and in terms of podcasting, yeah, I've been co-host on a podcast for over 10 years. Um yeah, do a video editing podcast, definitely. Um just love it.

SPEAKER_00 7:51

I'm always curious. It's it's kind of you know a cheap shot, but it's like, well, how much time does it take human versus AI, right? So you've you've been around the podcasting industry long enough to have done it the old way. And what what was what's the comparison? Okay, so the process you just walked us through with podcast AI is I upload it, I I change a couple settings, I approve a few outputs, but really it's a 30-minute process.

SPEAKER_01 8:22

Yeah, so so podcasting, it's really it's you know as little or as much effort as you want, right? So people can get away with like episode two, no description, no show notes, no chapters, no nothing. And guess what? You're gonna get no results. Right because there's no SEO. Um so if for the people that take the effort to do manually chapters on an hour-long video, that's maybe 15 to 30 minutes, depending on how good you want your chapters to be if you're doing them manually. Um, we take it down to 30 seconds, it's just automatic. Right. So just hit hit generate, it'll even detect the ads. It's it's amazing. Um the descriptions, they're they're really great. You just click the button. You don't have to think about it, you don't have to do anything, right? You just click. Our system is adaptive. It'll look at if you have a certain style that you do these in, you're coming in with a podcast, you're not brand new. Um, it'll actually be adaptive based on the style you've had in the past. So yeah, it's just super powerful. It's a major unlock. Um, it's kind of a no-brainer if you're running a podcast, really.

SPEAKER_00 9:24

But if you were, if you said, I don't need a tool, we've got our process, we're doing it internally already anyway, and it's working fine. How much time are those people spending on the equivalent output? A day?

SPEAKER_01 9:38

They're spending to do all that stuff manually or whatever their process is. Maybe they're chaining together a whole bunch of tools. Um, yeah, they could be spending like a great part of a day on an episode.

SPEAKER_00 9:49

So this is what's cool about AI, right? There's we've got all strata of business that listen to the podcast. Some of you have had an interest in thought leadership. Maybe you didn't at all, but you knew, hey, it would be good for my business if I did this stuff. But who's got a day to spend? And content is long play. It's not like a paid ad where you're going to see something and there's an instant result, right? So it's it's it's a a long, on ongoing effort. And if you have a tool like this, it's like I can do that on Saturday morning while I'm you know hanging out eating cereal or whatever, as compared to I'm running a business, I don't have a lot of time, there's a million things I could be doing, I'll get the content one day, right? So that's super cool that now any business or individual can uh leverage tool the AI-powered tools to be able to get a result in minutes rather than you know, and I guess you don't have to learn, there's no there's no skill set required. I mean, I'm I would imagine if like any other generative AI tool, it's pretty intuitive to use.

SPEAKER_01 10:55

Exactly. And and you know, that there's tools that are kind of closer to the AI, you feel more like you're more exposed to the limitations of the AI. It's kind of a lower level tool. So there are tools that feel kind of more like that. You know, we built a platform, we built a it's total, we're we're in the application layer, we're orchestrating a whole bunch of AI models behind the scenes, but for you, it's just a magical experience, right? It's kind of like using your iPhone. Your iPhone's doing a whole bunch of stuff, and you don't need to know how any of it works to just use it. And that's that's kind of the gold standard, right?

SPEAKER_00 11:32

This is a pleasant surprise. I wasn't expecting that. Um now getting back to actually using AI in the business, what was your initial exposure to it? If you've got a background in in programming and software development, those sorts of things, I'm sure you were uh using or um circulating with the the data side, like the data science side of AI, the ML, the like stuff that's been around. But as far as generative AI, what did that look like for you? When did that get introduced into your your biz?

SPEAKER_01 12:02

Um I would say the first time we looked at it was generative AI, uh was uh about the time of GPT2. And uh so OpenAI put out their public beta access. Uh we got access. And it was kind of interesting. We could basically say, hey, here's 10 titles of uh, let's say YouTube episodes, and it would keep generating more titles, like made-up titles. And okay, it was kind of cute, right? The the titles were plausible, you know, they'd be kind of made up, obviously, but kind of interesting. But you couldn't really make make something with it. Um the major unlock was obviously ChatGPT um about uh I'd say 15 months ago. And the funny thing is, ChatGPT wasn't even a technical revolution. It was actually just realizing two things that anybody could realize two years prior. And those two things that it realized was it so these GPT models, what they do is uh they're an autocomplete. They autocomplete one token, or you can think of it as a word at a time. And so you start typing, and it's just completing the sentence, what makes sense? What's the most likely continuation of the sentence based on the start of the sentence? And so it turns out uh realization number one, uh, if you start the sentence with uh you are an advanced AI agent tasked with helping a human, here's the human's first questions, how would you answer it? If you start the sentence with that, it will complete that story one word at a time, uh based on the premise that it's an advanced AI agent helping the human. That's not a technological revolution, that's just realizing how to use this thing. That's number one. Number two, second realization that made it possible. It turns out if you pass an array, just meaning if you pass here's the transcript of what was said so far in the conversation. Now it's taking the transcript of what was said so far in the conversation and the premise, you're an AI agent helping the human, here's what the human's asking. Those two things as the premise, it completes that, right? And that's what ChatGPT was based on. And you know, the reason why early on you had tons of hallucination was it didn't have enough information and it would kind of just make up stuff. But it turns out if you start putting into the prompt, hey, don't make stuff up, and you know, start putting in real information into the prompt, you kind of start solving those problems. And so Chat GPT is interesting because it wasn't a technological revolution, it was just realizing two little things that made an unbelievably compelling product 15 months ago, roughly. Um and that's when you know me and my co-founder we started realizing, oh my God, like it's kind of like you're in 1895 and all the parts needed to build a car technically exist. Someone just has to put them together, right? Like steering wheel, engine, a chariot. You can put them all together and you have a car, right? So the first person to start putting these models together could plausibly build, you know, an AGI. And by AGI, I don't mean, you know, super intelligent oracle. I mean like sort of a semi-replica of a person.

SPEAKER_00 16:12

Mm-hmm. It's in process from what they tell me. They're working very hard.

SPEAKER_01 16:18

Yeah, I mean, what we have right now, we have uh an ad read generator. You select which host of your podcast you want to read the ad read, you say who the sponsor name is, you put the call to action, and it will ad lib a two-minute ad read in your voice based on what it knows about your bio when you created that host in our system. Um, based on the background of the show, it knows what the show is. Um, and based on the call to action, what the sponsor name is, it will add ad lib the entire ad read, and it's pretty amazing.

SPEAKER_00 16:50

So the comparison, as I shared with you, like you know, we we built our own process with tools and APIs, and um, we're very proud of ourselves until I heard what you just shared with me. So the difference is is that that we yes, we've automated it, but we haven't really necessarily introduced a ton of intelligence into it. We use a tool that rewrites it, and then we use but so there is some basic stuff, but you're suggesting that in your platform, I can actually I can train my version, my my account. With those 1900 episodes that Jason had for his podcast, it got a very good idea of this is it very easy to autocomplete for Jason because it's seen all of his content.

SPEAKER_01 17:34

Interesting. Correct. So I I have a friend who um I started describing this to, and he's really big into sci-fi. And uh when I was describing this, he said, Oh, like a beta simulation. I said, What's that? And it turns out there's a sci-fi thing from 50 years ago. And a beta simulation is it has all the things that you put out in the world and to recreate a copy of you. And it's maybe 95 to 99% you know accurate, let's say. Yeah, yeah. An alpha, yeah. Now an alpha simulation is a molecular scan of your brain. That's going to be like 100% roughly. Um so the funny thing though with a beta simulation is okay, let's say we had everything that Steve Jobs ever put out and we put it into this thing. Okay, maybe it can get 99, 95% there. Um, but the difficulty becomes, okay, what is Steve Jobs' favorite color? Well, if he never actually explicitly said it, we don't know. But yeah, if the AI has this context, okay, what color were all his cars and what color was his computer, maybe it can actually infer what his favorite color was. Um so you really might be able to get to a 99% accuracy actually in terms of reproducing the person, which is pretty insane. So, you know, that's where the tech is going actually.

SPEAKER_00 18:44

Yep. So what once you saw, once you got access to to GPT-3, like 3.5, whatever, with like everybody else did, what was the did you immediately see the application for podcasting, or where did you where did you start? Was this like a straight path for you, or was there a lot of discovery?

SPEAKER_01 19:03

Well, we I basically started working on um recreating somebody on you. So um a business partner of mine um who's a cardiologist, sort of creating him as a a sort of an artificial cardiologist. Um the limitation was having structured output. So I'm gonna get really techie here. Um GPT uh OpenAI released something called GPT for functions. Uh functions means you get structured output, and you really need that to be able to do anything interesting, actually, because otherwise you're just getting inconsistent at random. It's very hard to deal with. It's unstructured. Um, so the thing that was holding back that was getting the structured output. Um, when that got put out, it obviously made a lot of things possible that are very interesting. But so started doing that. Then I actually created a parody of the all-in podcast. Um, so episode 117, uh, they mention, you know, uh David Sachs goes, Hey Jason, you're gonna get replaced by uh an AI. And you know, Jason says, No, Sachs, you're gonna get replaced by an AI. Um, and for those of you that don't know, uh, All In is the number one uh podcast, tech podcast. And um, so so I created a parody, like a 10-minute parody, and the premise was the reborn is AIs. Um, put out an episode on Twitter, uh, it got retweeted by uh Sax, the All In account, Jason went super viral. I ended up doing six episodes of this, so uh that was pretty damn successful. It went super viral. We kind of realized, hey, we have distribution for this. People were asking me, like, is it real? Um, is there a GitHub repo? And uh me and my co-founder realized, hey, we could make it real. So uh we so at the time, uh uh Press, who runs Jason Calicanus's uh founder university, which is his on-ramp to his accelerator, he was putting out applications or sorry, he was doing a call for applications to cohort five of Founder U. And uh so I asked him, hey, could should we apply? He was like, Yeah, you should definitely apply. Uh we applied, we bought podcastai.com, uh, and we got in and we started building about a year ago. Nice. Dang, that was fast. Yeah, we we um we went through the founder you program. We were one of the companies that got selected to have a follow-on, uh, like an investment, like a 25k kind of bookmark check um as a first investment. Uh, we got that around June. Uh, we got we launched and got our first customers in September. We got invited to a dim sum dinner uh event where we presented in um in New York City in September. And shortly after that, we got invited into the launch accelerator, so the 30th cohort. Uh, we went through that program uh pretty growing like really, really fast. Um yeah, we uh and foundry uh founder you and and uh the launch accelerator amazing programs. We ended up um uh going to Sequoia to pitch and all this stuff as part of the program. Uh we showed Roloff Botha. If you guys don't know, he's the guy who uh funded YouTube. He's uh the head of Sequoia, which is the biggest or most prestigious, let's say, uh VC. Um and uh we got ranked number one of our cohort. Uh and the way I did that is so Launch Accelerator, they train you to do this three-minute pitch. Uh 10 minutes before Roloff rolled in. I uh deleted the last third of my deck and I Googled what is his favorite wine. I found a 13-year-old core article where he answered and said, go to africanhunt.com to find a good South African wine. Found this random South African wine called Kontap Pinotage. And I and I loaded his podcast, Crucible Moments, which is produced by Vox. I loaded it into our system because I was like, oh, maybe I'll be able to give him a demo. But I realized just waiting there, okay, my only chance to demo is this pitch. So we deleted the last third of the pod of the uh of the of the deck. I generated an ad read of him doing an ad read for Conotap Pinota Pinotage for um for Crucible Moments. And so without even playing it, previewing it, I just got to the two-minute mark and I said, let me play this for you. So I played it. Um everybody's kind of floored, nobody knew this this was there. Uh Jason Kalkanston didn't even wasn't even aware we had this. And uh at the end, you know, uh Roloff, he goes, uh very good, but um it's pronounced Kinotal Pinot Pinotash. So that was pretty funny. Um he corrected the South African pronunciation. But uh he was he was quite impressed and put us in touch with their head of content. So um that was a that was a really cool thing. Um so yeah, yeah. So that was uh amazing and we were just closing around right now. And that's uh that's with launch tripling down on us. And so really uh amazing.

SPEAKER_00 23:47

That's fantastic. Yeah man. Hey so where where do you go from here with the or where do you do you better question do you have concerns about technologies that will be imminently released that may uh mitigate the the razzle dazzle of podcast ai and how do you prepare for that?

SPEAKER_01 24:11

Well we're basically we're doing that. So um it's it's kind of that thing where you know what did Apple do when they had the iPod? Well they killed the iPod. And what do you kill the iPod with? You kill it with um the iPhone. Yeah and so you know we're already we prototyped blog to podcast so we found people that do blogs pretty consistently we made it so it just scans their blog as part of the docket every Monday it generates an episode it ad libs an intro ad libs a conclusion ad libs all the the posts they've done into content and chapters um and it has intro music outro music it's fantastic.

SPEAKER_00 24:49

So when is that going to be released?

SPEAKER_01 24:52

So we have that prototype with uh with customers um if anybody's interested in using it for their uh for their blog we can do that right now um just get in touch and uh yeah our first customer who who started deploying it uh once we finished setting it up he asks okay so what do I have to do now right and I said nothing just keep blogging right this thing's on autopilot um so that's that's the first thing the second thing is uh so we just made two hires recently uh Greg Perkins who's our CTO that's to go even faster on the stuff we're already doing so yeah uh building in push to YouTube uh building in the clipping stuff in the cloud building in connecting all your social media accounts to put out that those uh social media posts automatically on top of it all in one platform so that's gonna be a massive uh thing for us and for the platform we also just hired Sean Allen who's a really prodigious uh iOS and and all all Mac platform uh developer and engineer and he's gonna be building a consumer podcast app and uh if anybody in your audience are trekies and you're aware of what the holodeck is um get ready because that's coming and that's the future of podcasting and we're building it. So the answer to your question is we're we're gonna obsolete ourselves.

SPEAKER_00 26:08

Yeah man that's so cool I can't wait to turn this tool on I was for everybody listening I was looking at the wrong side I thought Edward was from a different company. So I had done my due diligence my research on the wrong business. So I'm learning all this cool stuff right now just like you guys are and I hope that my content team when they get this takes the hint let's test this out. This is really cool Edward man this is fantastic. Thank you. Now as far as one of the things I always like to ask is how you're using generative AI, not as your business but in your business. Any clever ways that you're using tools uh for you know use cases that might be familiar to others?

SPEAKER_01 26:49

Yeah 100% so uh first of all for coding chat GPT is super helpful right um that eliminates stack overflow hours and hours stuck on stack overflow trying to find the answer um chat GPT won't solve everything but it's actually super helpful to unblock you or lead you in the right direction that's number one. Number two, if you're actually coding um Visual Studio now has these uh you know Microsoft Copilot built in um you can use an extension for that super helpful um again it won't write everything for you but it will uh essentially guess based on the context what you're trying to get at and you know it'll save you a good chunk of time so that's an amazing tool as well. Frankly ChatGPT as an LLM is useful for prototyping product. So what are we doing with podcast AI? We're we're application layer so we're orchestrating a whole bunch of models together to do something interesting. You know what are you Chris what are you doing? You're showing how to actually do this as well from a technical perspective plugging things together. I suppose the difference would be we're approaching it as let's build a product right so that we're maintaining this stuff. We're making sure the plumbing works right and now you get to use it off the shelf you don't have to understand how it works. For you it's just magic and that's the way it's supposed to be right your phone is magic. You don't need to know how your the gyroscope and the accelerometer work it just works right so so you know we use ChatGPT actually sometimes to prototype prompts uh and and you know the first prototype of of a product actually or feature sometimes happens in one of those LLMs and you guys have access to those LLMs you can do the same thing. You can figure out what sequence of prompts gets you a cool result and you can keep manually doing that. And but the cool thing is that can actually become a product or that's the way you prototype a product. So that's a that's another way to do it.

SPEAKER_00 28:46

So tell me about the the the prototyping this is so normally we're getting uh people from all all careers but this is like episode 20 something so it's not like we've had uh the widest spectrum of of the the working population we haven't had a lot of tech folks that are that have a background in programming and understanding technology's application in the real world and things like that. So prototyping is not something that I've done can you maybe talk a little bit about how you do that using the models?

SPEAKER_01 29:15

Yeah so so when you're creating uh an application what you're doing is you're you're uh orchestrating these different tools and models uh but you're orchestrating in a way that's going to be repeatable systematically so uh I upload my episode I click the generate description and I will always get a description of my podcast episode and how that works is in the back end on our API which is basically the application you're using talks to the back end and I says hey I need a description for this episode and our back end is has a prompt you know it has a system message it has a user message and it sends it to the LLM and it gets some structured output and then it takes that output in a structured way puts it in the database and and sends it back out to the front end where you're consuming it. So that's all structured and everything. But to get to that you have to figure out okay what prompt is going to give you the result that you want and how do you prototype that well you could prototype it just in code but the first step actually to prototyping something is just go to ChatGPT and write something that you think can give you a systematic output based on your input. So um you are an advanced uh cook uh I would like a recipe in bullet put bullet point form um for the for this type of cuisine uh the this is the list of ingredients and they're comma separated uh give me the bullet put uh bullet point form uh recipe uh including duration at the end uh write the duration comma space you know the amount of minutes so that's your first attempt to get sort of a systematic structured output right if you're writing an application that's what you would need as the output right so pretty quickly if that looks like it's kind of working we'll now take it to code and we'll start doing the the the development in code but the first prototype is just in chat GPT what can you get?

SPEAKER_00 31:21

Yeah okay now that makes sense you know man i'm I'm sitting over here and I realize that if I was in um Bangladesh or uh Namibia or anywhere if I had this tool I could be a podcast producer I could do I could use your platform as the engine for you know me hanging a shingle saying hey I'm I'm a podcast producer and going out there and soliciting business I don't even have to do a podcast to monetize your tool I can use your tool as the fulfillment side of me as a service right like like podcast production as a wow very cool man it's Edward I'd love to be able to support you guys and I I'd obviously I I'm encouraging the audience to go check this out even if you're not podcasting here's what's happening is it's a lot easier to create content now. There's a lot of crap content that is being created because people haven't taken the time to um get comfortable or familiar enough with how to communicate as Edward was talking about earlier one of those two breakthroughs that had nothing to do with technology it had to do with how you communicated with the tool how you used the tool so I would encourage you even just as a well I'm not gonna podcast Chris that's not for me. It's not about that it's about you learning how other people are translating idea through or powered by generative AI which is not now Edward's doing a level above what most of us will be doing however um it's it's a it's a good opportunity for you to see how what's possible when somebody gets creative with these tools. You can get creative with these tools. The tools themselves aren't that hard now if you need some technical expertise you find the up and coming budding Edward somewhere on on upwork or whatever but uh and if you are thinking about generating content I can tell you that shirt writing some people love doing it I do it I communicate but I'm not a writer podcast was easy for me. I enjoy talking I enjoy learning and meeting other people new people it's it becomes now this this as you noticed you recorded the video on Zoom or Loom or wherever you did it you uploaded it to the platform but now it's also creating stuff that can coming soon automatically be posted to your social profile. So anytime that you've got positions on what's occurring in your marketplace or your industry or with your clients if you take a few minutes and record that you now have dozens of pieces of content that can go you know across multiple platforms and it took you the time for you to remember oh yeah I need to be I need to let people know I'm an expert at something it's beautiful.

SPEAKER_01 34:09

Well man great surprise I'm very glad that this uh Edward and I had some he was traveling I was uh busy in startup land so this almost didn't happen but I'm glad that it did man so for people that want to I don't know obviously they can go to podcastai.com and of course it'll be in the show notes but is there any other resources or or place you want to send somebody so that they can uh learn more about what you guys are up to or follow you guys uh yeah uh follow us on Twitter you can follow me on Twitter uh at uh Edward Brouwer um you can follow podcastai on Twitter at get podcastai uh and again podcastai.com that's the website and uh whether you already have a pod or you're just starting um yeah we're we're a tool for everybody um and we have some of the top pods out there we just signed on the number two uh pod in medical last week so pretty excited congratulations yeah man thank you you guys are gonna be taking over well I think you got a new client here I think I think we're probably gonna test your stuff out in the next 24 hours so awesome and thank you so much for being a guest on using AI at work man and again um uh I'd love to follow what you're doing I'm gonna follow you guys on on Twitter and and and watch the progress it's future's very bright for what you guys are doing over there cool man uh it's been a pleasure Chris and uh it's great to have shows like yours out there to get people into this because this is this is amazing this is like the internet starting 30 years ago this is the greatest opportunity of our lifetimes super fun get involved folks all right we'll see you on the next episode thanks for tuning in to using AI at work don't forget to subscribe for more conversations about how to use AI at work and a special thank you to our sponsor Chief AI Officer for empowering businesses with AI education and training.

SPEAKER_00 35:59

Visit their website for free AI readiness assessment and AI strategy guide to help you get started using AI at work. That's www.chieofficer dot com thanks to our producer Evan Despolnier for making this episode possible. Follow us on Twitter at the handle usingAI at work and visit www.usingai at work dot com for free resources to help you harness AI in your role