How to monetize your content with AI, with Arman Assadi

In this session, we will talk about how to unlock knowledge from inaccessible content. And I'm really excited to have Arman Assadi here. He's the founder and CEO of Steno.ai And I think I've known Arman for I believe 10 years now. We did a session just like this kind of, when he was just quitting his job at Google.

And that was around the time when I quit my job. I don't know if I hosted my first summit. By then, I believe I might have, and it was just afterwards when I started traveling, I think I was in Cancun, and I had this background, I remember this, it might be even on YouTube right now, so you can check it out, if I find it, I'll be sure to link it up in the description, but yeah, Arman has done a lot of things since then as well, launched multiple companies, you know, wrote copy for some of the biggest names, and I just see him, His success is really great to see and have him on for this session here.

So warm welcome, Arman. Thank you. Thank you, Navid. I appreciate it. It's crazy to think it's been a decade. hopefully we've grown a little since then. No, I'm sure we haven't. Yeah. Tell me a little bit about your way into the world of AI and maybe a little bit of background if someone doesn't know, more, about.

Yeah, sure. So what pushed me to get into AI was, one of my mentors is a guy named Kevin Kelly. And, Kevin is the founder of Wired Magazine and an AI futurist. He's got some incredible books, What Technology Wants, The Inevitable, all about AI and what the future holds. And I've been really fortunate to have interviewed him a few times.

And one time we were sitting in his library, in his home in Northern California, and we were talking about one of the concepts from his books. And he was telling me about the exponential explosion and growth of content, video and audio content, and how critical it would be to organize that information and unlock that knowledge to actually make it useful for people.

And that if we did not do so much incredible content like this would be lost. The best information today is being published in audio and video format through podcasts, YouTube, social media, so many platforms, right? People rarely these days wait five years to write a book or publish an academic paper.

Even the greatest scientists of today Frequently show up on podcasts and share their information and their wisdom. And so there's a big job for AI, there. And that really inspired me. in, 2020, when the pandemic first started, I, like everybody else, was having an existential crisis and sitting around and thinking about what to do and how to spend my time.

And I was sitting with one of my really good friends, Nick Urbani, who I've known since college, since SDSU days. And, We were just looking at his Asana and he has this Asana task with like business ideas in it. And we were just like, yeah, let's see like what's sitting there. And one of the ideas that was in there was exactly what I had been thinking about for.

At the time it was, it was 2020, it was 2016 was when I had that conversation with Kevin Kelly and I had been thinking about it and planning to start this company for four years. And he had this thing, it was like, you know, something like a Wikipedia for podcasts. I was like, holy shit, dude, that's, exactly what I've been thinking about.

And he was like, if you want to build it. I'll cut you a check right now and be the first investor in the company. And I was like, what? All right, let's do it. so I did and, started it in 2020 before AI was where it's at today. obviously AI has evolved a lot and things were a lot more difficult.

There were no easy to use APIs and. No OpenAI and ChatGPT doing its, there was, but it wasn't what it is today. It was around the time when Jasper and Copy.ai came out. That was the first, like a lot of marketers started using it. That's when I first dabbled with, generative AI. And, but now it's of course a lot easier to access it with ChatGPT OpenAI's API and things like that for people to build on top of it.

And yeah, that's right. Yeah, that's right. No, that's awesome actually. what else, what happened I 2020, you kind of got into that and then Yeah, I got into it. I was, you know, I, just jumped in wholeheartedly and once I did, I really started loving the space and one of my. Huge life goals, has been to, eventually get to a point where, as you mentioned, we've done a lot of different things over this last decade.

I had an e commerce company. I built a mobile application. I did a lot of consulting and copywriting for these big names. And that was really fun. I really dove into the world of digital marketing. but I always loved products and I always loved technology. And as a CEO in my former company, I really wanted to do a B2B SaaS play.

I really wanted to build a proper startup. And I had always in the back of my mind said, I've done the bootstrapping thing. I've done the crowdfunding thing. I want to raise venture capital. I want to really understand that world and jump into it. And so this has always been a goal of mine to build a technology company in the B2B SaaS space.

It's been a journey over the last three years of just like being in it. I finally jumped in full time about a year ago and it's been just nuts, just nuts. It's just a whole different world. It's very fast paced in the AI world for sure. So Arman, so for people who They don't know what Steno does.

I mentioned a little bit in the interview, you unlock knowledge from inaccessible content. I guess explain that a little bit more, because you started, you started a little bit different. Now you're kind of expanding to, that's right, more to events and things like, kind of like summits as well.

We talked about that a little bit before in the pre chat, but. This is just a brief explanation of Steno and what you're, where you're at. Yeah. Yeah. So as you said, you start one way and as you start building and talking to users and talking to customers, things move and you do that early stage customer discovery.

And these are the things that I love about tech and I love about the education that's out there. And I have some incredible mentors who, honestly without them, I don't think I would have figured this stuff out. And so as we went down that road, we put out a great product, we got to the point where we had 25,000 monthly active users with our first initial version of our product that was focused on podcasts, making them searchable, providing a chatbot to be able to ask any question and we scaled the shit out of that.

And it was awesome. But along the way, there was always deep down for me, a question Is this the right market? And is this the right customer? And, I recently joined Techstars. So we're a company that's now backed by Techstars. For those that don't know, Techstars is the largest pre seed investor in the world.

Something like 3, 700 companies have gone through. And, it's got a bunch of unicorns as a result of that. and very illustrious, very similar to Y Combinator. These are the two big dogs in the And as a result of that, we stepped in and I had this question deep down inside of I really want to figure this out.

Is there potentially a pivot that needs to be had here? And we very quickly found out that there was. And with that comes a tremendous amount of pressure to get things right and to move quickly. But also this incredible, just a feeling of confidence as well that you're beginning to move in the right direction in the journey of product market fit.

the way I describe what Steno does today, we help media companies and professional creators. We take all of their content across the multiple platforms that exist, we bring it all together into one space, one platform, and we make it searchable, first and foremost, and then we give their audience and their users a chat interface to be able to ask any question.

For the, company, for the creator or the media company, this creates a whole new revenue stream because We basically, we're a B2B2C play, so the audience comes to Steno, pays 10, 20 30, whatever the company decides, a month on a subscription to access all of this content in one place, and we do a rev share with the company, and the company makes a whole new revenue stream, and for the user, we've already learned that people want this stuff.

Just for example, from podcasts, what we learned is you're overwhelmed. There's so many episodes, you subscribe to so many different podcasts, and you don't get to keep up with it. And sometimes you just need to be able to search, and what did Huberman say about that supplement for sleep?

What is his sleep stack? There's so many examples of that. Or, this episode looks interesting, but I only care about this part. This podcast covers seven different things in this episode. So by making it searchable, by transcribing it, creating summaries, we chapterize it, we create translations as well for accessibility.

This all just makes the content more and more accessible to more people. As a creator, that's what you want. that's what we've learned. You spend all your time creating content and media companies even more so. It's about distribution. So we're helping them distribute their content, and get it in front of more people.

Yeah, I think that's possibly what you do also differently because you mentioned something about the rev share model. I haven't really seen that. With other AI startups or people doing, I mean, there's a lot of, tools where you can actually upload your knowledge, you can build it yourself, but there's, there's no code tools where you can actually build, you just upload, let's say your blog or, you know, your videos or whatever that is, and you can have it searchable.

But I guess it's not like an instant, you know, monetization, like I, tell me a little bit more about that. I think that's interesting, actually, that you, you're going to like work together. It's like, it's like a collaboration in a way. Yeah, it's a partnership. That's right. That's right.

It's a partnership, maybe to make it apply as well to people that are thinking of starting something or people that have built tools, when you are trying to attract a new customer. Especially in the early days, in the early days, you just want to do a pilot. You just want to see and learn.

And so by doing these pilots, what we're learning and have learned is It's all about making it the lowest friction, lowest risk, lowest commitment way to get your product out the door in front of companies and to get embedded into their workflows. So we realized very quickly that for many companies, we're creating a new revenue channel.

One of the areas that we were focusing on well after podcasts, as you alluded to, is conferences, virtual summits. Those recordings that exist, especially for conferences that are in person. There's seven talks happening at the same time. People miss those talks. And many of the conferences we go to, they sell us the recordings afterward, right?

Anywhere from a hundred to, I've seen up to, 500 to buy these recordings. You buy the recordings because you're like, I might need these. But then they just sit on your computer and you never go through them. it's, again, it's too much content. It's overwhelming. So by partnering with the conference or the virtual summit, we're able to provide these recordings to their audience in an AI powered way that makes them searchable and accessible so that the user can get to exactly what they're looking for.

And by doing a rev share. If the company decides to sell this access to this content, whatever they decide to sell it for, and we can handle the sales process for them as well, we just do a rev share. like what Apple does for an app, taking a percentage, taking a cut of the revenue.

And that opens up a lot of doors, for Steno. And I think it's an interesting, approach to, to go to market. For sure, it's way different than what I've seen because I'm going, if I see on your website, it says AI powered search model, right? It says transcript and summaries, it says instant translations.

That's pretty interesting to get in front of, a Spanish speaking audience and then the chat. If you go over a little bit, is there anything that you have done since, I know it's evolving very quickly, even like when we talked like a few, it was like a few weeks ago, you changed the sales page very quickly.

What's completely different? So I don't know if there's anything else, but there might be something else you have done since then, actually, that you're adding to it. Yeah, it's, there's a ton. I'm really fortunate because my co founders are crazy, like me, and they move really fast, and they're hungry and motivated.

And they're technical as hell. And, one is a master's level data scientist and back end developer. The other is a master's front end developer and to, together, they actually won the Polkadot Hackathon out of 1100 projects. So they're just, they're machines. And so we're always talking to users, finding what's next, and building, in the way that we know best.

And so there's so many. Features and exciting things on the road map. one example, just even what you and I were talking about before we started recording is once you have access to that content from that company or that creator that you really care about, just the ability to save those insights, right?

You start to highlight things. That's a feature we have. You start to highlight. You start to capture. There's so much that we can feed back to the company or the creator about the way that their audience is interacting with their content. For example, if everyone's asking the same question to the chatbot and the creator thinks that what these people really care about is like marketing, but the question they keep getting is more, product related or, business development related.

They keep asking questions about business development or personal development. We feed all of these analytics back to the company. Back to the creator. So if you think about the ways that you currently learn as a creator, what people like. You're just relying on comments, on analytics from, your Instagram, your YouTube.

You're not getting a direct insight as much as you would like, because it's not based on people can't see as much, they watch time and things like that. This is a whole new way of understanding audience engagement, the questions they ask, what they're highlighting. You know, when you're Kindle, you see like popular highlights, like that's the section people keep.

Repeatedly highlighting, same thing, being able to feed that back. and for the user, what's really interesting is once you start to create this like portfolio, this hub of all the things that matter to you in the world of content and what matters to you in your work, in your business, or whatever it might be, giving those users tools, AI tools, to be able to then turn Okay, take all of my learnings from the AI for Creators summit and turn it into something.

Turn it into a presentation for me to share with my team. Turn it into an executive summary. these are basic things that are now very doable and that's just this stage. But the stage that we're really evolving to, if we want to get out there a little more is Automation, right? We had a snippet of that earlier this year with agents and these agents that, everyone's throwing around the phrase like AI copilot for this, AI copilot for that.

And that's the next stage. That next stage goes even beyond a copilot into an autopilot, right? Where I believe that the trend toward AI needs to be adaptive. And personalized to the individual user, creating an experience for them that is really nuanced and personalized for what they need, so that when you then come to the platform, if you come to Steno, for example, what we want to be able to do soon is, hey, we already know what you need, what you like, the exact insights that you're looking for, and we're going to start to source that stuff for you.

Without you needing to go and manually read or search or ask a question, and that trend is going to start to explode everywhere. Across every medium is being able to just feed you what you want more quickly. No, this is really, really interesting. And I think I think something else, I mean, maybe you have already thought of this, but like, I'm not sure if you're integrating the, let's say the experts.

I mean, if you have a summit, you have multiple experts, but for example, what about voice and maybe, you know, cloning avatars and things. I don't know if that's something you have, I mean, 11ElevenLabs and HeyGen and things like that, but that could be something. Potentially to get into, so it's not just text based essentially, you know, that, that could be a cool avenue.

I don't know if that's something you're thinking about or that you can see this going. Definitely. Yeah. Yeah. I think that creating a multimodal experience is very important. at the time that we're doing this talk, OpenAI announced that's coming to ChatGPT. For me, it hasn't released in my ChatGPT yet, but.

You won't need to switch, modes anymore and you're going to be, yeah, it's going to be just like one experience of like audio, video, text, image, video, not really yet, but image and soon video, generate me a hero, a one minute hero video. On everything we just discussed, I mean, how cool will that be?

And those foundation model, achievements will carry over to the application level. again, like ChatGPT is just an application that is. It's showcasing what the foundation model is capable of. It's just an interface. I don't want to repeat anything people already know, but just making that clear.

And so what that means is that all the AI startups out there will be able to come up with their own creative ways of showcasing what these foundational models are able to do. Not to mention that things. like fine tuning your own model become easier and easier to do. And so there will be a lot of unique moats, I believe, like the moats are not in the data.

people have this like constant thing in the AI space with how are you defending what you've built? How is this defensible? ChatGPT, OpenAI is gonna just wake up one day and go GPT, ChatGPT7 is just gonna crush it. That's true in many cases. As a founder, you have to be thinking ahead and you can't be naive about what you're building.

You have to make sure that you're thinking about defensibility. But at the end of the day, I don't think the defensibility is in the, data. I believe it was Sequoia came out with a really great paper on this and to paraphrase it, it was something like foundation models will obliterate any data moats that exist.

And so ultimately the moats are in the customers, like your ability to get distribution customers and create a unique application workflow. what is an existing manual thing that people do? Call it a workflow and automating that and eventually getting to the point where it happens for them.

That's the moat, man. the faster you do that, the better you do that, you win. And, yeah. No, I love it. on that, I just thought of something like, for example, if you're a creator and Let's say you're publishing and you have a lot of podcasts, you have a lot of videos. Now, of course, someone might push back and say, there's a lot of like summary tools and there's a lot of like things that can do this for me.

what do we say like for using this model that you are, you're pioneering it's then, or what would you say to them to get over and using that to maybe monetizing and things like that? Because there are of course, tools that compete. I just want to make sure we, we speak to that as well.

Yeah, of course. And everything comes down to cost and time so that's what it's about. Like right now, if I want, and again, these tools change every day, there's always new ways to do something. for example, let's just use like what we used to do heavily on one. Imagine all you were trying to do was get one podcast.

And do what we do, right? and again, not exactly what we do, because there's so much under the hood to what we do. I guess you're automating it, so you basically take the podcast feed or the, I'm sure you're doing it automated, so you don't need to take every link. Every time you're publishing, you need to go there and do it, or you're taking this feed automatically.

I assume that's something. Yeah, Steno is doing it for the publisher, and then the user just shows up and logs in. And the moment the episode went live, it's there. But if you wanted to manually do this workflow, and not to mention there, it's their courses, it's their YouTube, it's their summit talks, it's everything.

But if you wanted to manually do this as an individual, by all means, go for it. But think about what you'd have to do. You'd have to download the file. You'd have to go and get it transcribed, which is not. It's getting cheaper, but it's not that cheap, right? Descript and stuff do it, pretty accurate.

10 hours a month, 10, right? If you're a podcaster, you might be using it, but either way, most people are not using, I feel most people are not using, Rev.com or you can use even YouTube's transcript.com gets you a transcript, but it's not, maybe not the most accurate. I haven't tested everything with this.

So that's right. That's right. So no, yeah, you got it. It's if it's on YouTube, you can get the subtitles, the captions, right? They're not the highest quality. They'll get better. Of course they'll get better. it's Google, so like they'll eventually get much better, but say you wanted really high quality ones.

So you go to Rev. You pay a quarter, 25 cents a minute, so if you have a 60 minute, podcast, that's 15. Okay, so that's interesting, Now you have that, and you take it, and you go to Claude Can't go to ChatGPT yet, because the context window is too small. So you go to Claude One of my favorite tools, by the way, is Claude And by the way, they're crushing.

Google just gave them 2 billion. Amazon gave them 4 billion. It's getting wild. I use ChatGPT a lot. I have other specialized tool. I'm sure we can get into that a little bit. Yeah, we should get into it. different ones, different purposes, right? If I want to upload a legal document that's 90 pages long, I'm going to Claude.

I'm like, here you go, Claude. And then I go, okay, summarize this. Cool. That's fine. That's easy. And I can chat with it. But this is a, one episode, four steps, 15 operation. And so the way you win in AI is you take, like I was saying earlier, you There's manual workflows, and you automate them, but then you also think about how to do them at scale, and you think about how to do them in a way that differentiates your, the way that you do it.

And so there's so much that we do, like speaker detection, right? We actually do a lot of interesting things under the hood, even with just a video, to be able to use computer vision to detect who that speaker is, even if the name is never said. In the talk, say we took this video and you never introduced me, a lot of podcasts don't really introduce to get, themselves if they're co hosts, like we don't on my podcast, Alfalfa.

Yeah, I didn't say my name in the beginning here. Exactly. Yeah, this is interesting because in Descript, which I use to edit a lot if I don't need like final cut or something more advanced. Basically, you can, the transcript is automated. It's actually pretty high quality. What my process, and then it can detect the speakers, but you will have to, you might have, I haven't checked that, but I think I have to enter the name manually, but they Yes, I'm talking about entering the name.

Basically, by my voice, if I put Navid speaking, then they will know entire conversation that I'm speaking. And the same with you, so that's what they are doing, but automatic detection. I haven't seen if they are doing that for them without me actually saying it in the beginning.

So that's right. That's right. So look, ultimately it's a, it's like, are some of the use cases and different tools similar? And could an individual user do this for a small amount of content? Yeah. But we have. A ton of GPUs, and we do this at scale for these companies who want to put hundreds of hours of content, if not thousands of hours, that is a better explanation, in front of their audience.

And it's instant and it's constantly getting updated at scale. And so like an individual person. If they wanted to do that at scale, it's virtually impossible for now. yeah, and then also speaking in terms of as a startup founder, like we're, we've been discussing the use cases for a content consumer, right?

A user, but as a startup founder, I don't, I'm not competing on that. I'm competing on the incredible amounts of. There's a lot of data that we are accumulating as a result of this, being able to, for example, source everything that Navid has said across every medium, every channel, and making that useful.

Does that exist today? No. And so there's a lot of interesting possibilities when it's like, Lex Freedman. Lex Freedman has his own podcast, but he also is a guest sometimes on podcasts. Sometimes he might go on, I don't know, CNN and talk to somebody. Is that something you can do?

You can take, let's say, a famous, I mean, it's more, it's easier for a famous person, but even any creator who has a lot of content out there, you can basically see. Outside of their own channels, you can go and say, okay, he was on Forbes. He said this there, and you can basically get that in there.

So let's say someone wants to chat with, Steno the Navid bot in there or whatever, you could actually get all the information they want. that's the deeper, longer play about, it's about the data. It's about developing a proprietary, unique data set. And God knows who knows what humanity will want to do, what companies will want to build with these data sets.

That's ultimately what OpenAI has done. They've developed and cultivated this huge data set. And then companies build ChatGPT wrappers. the ones that are doing that, we don't do that, but there are many companies that do that, and they're building useful tools with that data. The companies that are doing a little better are the ones that find or build a proprietary data set.

In order to really compete in AI, I think that's part of what's important. I still think what I said earlier is the most important. It's not only about the data, it's about the distribution and about getting users and making workflows faster. But if you can get a proprietary data set, You have strong defensibility.

Yeah, really interesting, but going back to the way creators can monetize or adding an extra revenue stream, because I think that's where it really makes sense. Things like that, and also the data is really important, like you can get all this data from people asking questions and things like that.

that's amazing. But like in terms of monetization, is that something you could like embed in your own? Let's say you have a membership where you have your customer base and things like that, or do you need to send them like the standout to go in there as well to log in to get this information or is that something you're working on?

I'm just trying to understand how this could work for the average creator who's doing courses. Yeah. And let's discuss like how much money. potentially can be made from this revenue stream is one of the keys is you don't want to cannibalize the existing revenue streams you want to add, right?

So let's just say you have, your first question about sending people to your Steno page versus embedding it into your page. These are the two options that we currently have, and obviously the option of integrating and embedding is more expensive and more difficult, there's some onboarding expense there, but over time that's going to get a lot easier and a lot faster, and our integration team will be able to do that in a much more seamless way, the way that you just embed it.

Like, you're a site builder. You've been in this game for a long time. The way you take a plugin and embed it into WordPress or something. Like it just, it's going to get to that level. I was just thinking like you have your community somewhere or you have like your. let's say you have a really amazing membership site, you want to have your branding, everything in there, and you embed this into the site.

I think that's a really cool feature. And I guess you do it now, but it's going to get easier. It's going to get easier and faster. And there's, what we've learned from creators and media companies is there's a tremendous value in it. Being not just the courses they have, but the courses being one collection, one component with all of the free content they have, the premium newsletter that they have, all in one place, making that all searchable so that people can chat with everything.

Because the main problem that's being solved for your, as a creator, your audience is like, it's not just your course. People love your content. They just can't keep up with everything all the time. But if they could have everything in one place, it would make it a lot more valuable for them.

So both options exist, but let's just say you were a creator with 10,000 people. And, in marketing, we always say, a decent conversion rate is like 3%, if you get 300 people to pay you and say that you decide you're going to have a 10 tier Which is for all your free content, then you're going to have a 30 tier because that's the content that is normally premium for you, or is your course content or whatever, and you want that to be more expensive, right?

But let's just say the 10 a month, you're charging 10 a month, super small amount, that's 3,000 a month on, that's 300 people, that's the 3%, 10,000 10 a month, that's 3,000 a month in additional revenue and say Steno takes a percent of that because we do a rev share, but you're basically looking at over 2,000$ a month.

That you're generating on a small $10,000 person audience. So you don't take too much percentage. I don't know if you worked out everything at this year's, correctly with some conferences and stuff like that, or have you done, have you started doing it for course creators and things like that as well?

all of the above. and it just depends. So it's just a matter of getting in touch with us. like figuring out what works for you, but the majority of the money is yours. I'm a creator, right? I have my own podcast and I want the majority of my stuff to go to me. That's a really awesome model because you're taking care of, like some people, there's tools, as you said, you can do every, everything exists to set up different things yourself.

But maybe you love doing your YouTube, your podcast, but you don't want to do the rest. You just want someone to take care of it. you can use a platform like self hosted WordPress, you set up everything, right? It's a lot more work. Maybe it's going to look more branded and more professional, but then you can also go with selling on Gumroad or something like that.

But they're taking like a 10 and it's maybe not going to be, it can be really good for some people, but at the same time, you're already doing it. you're already sending people to YouTube. You're already sending people to your OnlyFans, whatever. and OnlyFans isn't just, corn anymore, right?

there's people putting actual content and utilizing it exactly how, think of it like OnlyFans, you have subscribers. And OnlyFans is taking a percent. So, like It's a funny analogy, like, but it is like that. I mean, I wasn't sure what kind of percentage you were talking about. So, are you, is it like We're the minority percentage.

Is it like free to sign up for it? If you have like content, or how does it work, really? I'm just trying to We will soon have yes, it's free for the creator and the companies, it's free. And, we will soon have it where you can come to the site and like self service, software, like here's a link to my YouTube.

Here's a link to my podcast. Here's a link to my course. And then our integration person will just reach out. Hi, what's up? How can we make this Beautiful for you, blah, blah, blah, but for now, yeah, you're just getting in touch with, the team and we're doing it, we're making it happen.

But yeah, there's, that was my point earlier, creators work so hard to create so much content and you and I, of course, know, and many creators out there know about the 80 20 rule, right? 80/20 rule says that you should spend 80 of the time promoting your content, not creating it.

But what's the reality? Maybe even 90%. yeah. But what's the reality? You spend most of your time thinking about the content that you're going to create. 95 creating. It's fucking opposite. And this is a way to get your content out there without any extra work and a way to monetize it.

Creator economy, I don't even need to tell you, is so huge, right? Huge. 250 million dollars. There is no other global economy that will be half a trillion dollars in the next three years. There are 200 million creators out there today. And of those 200 million, by the way, about 50 million creators and media companies, media companies, of course, make a whole time living, but of the creators, 50 million of them are making, excuse me, not 50, 5 million of them, 10 approximately, are making, full time income, like they're living off this stuff.

So if they can find a way to make more money and help their audience consume their content more, yeah, that's why they like it. I'm just thinking kind of what this can replace it's going to replace some certain things, right? Because Patreon, for example. It's pretty, I feel it's not the most useful.

I know some creators use it, but I feel it's quite useless sometimes, to be honest. I haven't used it, but I see some creators doing it, but there's better, I think, I feel there's much better ways to monetize your audience than a Patreon or something like that, where you're just giving.

You're also giving them a percentage, by the way. It's a similar model, but I don't feel the, it's not always the value. I know some people show up like for a Q& A or something like that, but I feel this model, It's better because you have a lot of content. Like imagine, let's say, I think a big creator or marketer, like I know you've been working with him before, Neil Patel.

He has like crazy amount of content. Imagine he, you can go to his blog, you can go to his YouTube channel, but you're not, probably not going to find everything, right? I don't know. I don't know. I think he has his content that's searchable and things like that. Yeah. Cause semantic search as well as what we have.

Semantic search for people that aren't familiar it's a better search. It's the ability to just show me you speak in normal language, show me everything about, content marketing for small businesses, and being able to ask a question like that, is really powerful. Neil's a great example.

Neil's a great example, by the way, 15 years of dense blog articles. I think there's quite a few, but he's one of the best examples for sure. he's just a machine when it comes to producing content. And now he's more into podcasting and videos, of course, but there's so much content out there.

And I feel like a lot of people, they also take. Like for paid products, I've seen like books, they're basically like a collection of the blog. I remember like Michael Hyatt back in the day, he was like, I think it was like platform, one of his first books, right? And was like, he took it from, so like collection, I mean, probably even he said that, I think it was like a collection of his blog posts that he expanded upon and things like that.

And he took it to his book. So it's not a foreign concept to actually take even what's out there for free. In a way, and making it more convenient to people and sell. Yeah, my prediction though Navid is I don't know, eventually I think again, these, this world moves so fast, but like I can see a play where, we do become that hub where you initially upload as well.

it isn't just necessarily. Capturing and bringing it all to one hub, but it's like Kajabi, imagine like Kajabi has turned into a multi billion dollar company and a great success and, like Patreon, that's an interesting way of looking at like the competitive landscape and seeing okay, Kajabi's currently focused on course content and being able to publish that in a beautiful way and put that out there.

But, I could see a play for it also being the initial base for where you Publisher, course content in particular. I think it's so interesting because a lot of companies, they start with one specific thing. Kajabi, as you mentioned, this, I don't know, they have been like 10 years now.

Very long, they had an early version. It was only courses, by the way. It was not like, now they're like an all in one tool. Which is not the best at everything, but they are doing some, the course thing they do pretty well, and the rest they do okay, I would say, they do it, but not maybe as the best as the specialized tools, but I think, for example, for Steno, I guess you could, because you're gonna be so focused on getting all this searchable content, you can probably create like a platform at some point that people can actually post their work, Courses or paid content in a way that's, accessible to their users.

so they can have their branding and things like that and that's something I could see happening actually. And that's, Yeah. And you're talking about some really great, just like general startup principles, right? You build for the thin edge of a wedge into a market, and you do it really well, and you go as niche as you can, just like you do with content, right?

You speak to a niche audience, and you build for a niche audience, and then you allow that to organically spread. And when you achieve product market fit, when you know you've achieved product market fit, is when all these people are like when you have too many sales calls, when you literally can't keep up anymore people are hacking at your product to try to get it to work for them because they're so excited about it and your market is expanding and your emails are exploding and everyone wants to talk to you.

that's, product market fit. And at that point, like you can begin to expand, the market you serve a little bit. And as you ask the questions of like, how can I help and what can we do, for you, I think we're going to learn and a startup would learn, okay, this is definitely the next thing to build.

There are frameworks for this, by the way, that I could mention, but they're, like Superhuman, the email app actually has a really good. Framework for product market fit, available. I love that app actually. It's really good too. Yeah, me too. But that's what they've done, right?

They were initially serving somebody very different. they've expanded that. You couldn't even get access to it without an onboarding call. Didn't even that before. I just found it. I don't know. I think I found it like this year. It reached you. Yeah. It finally reaches you. and it's usually word of mouth and people be like, Oh yeah.

Hell yeah. Superhuman, man. You better try it. So like Ali Abdaal mentioned it, the one that he was using it, he's all into productivity, right? So I was like, okay, this is the best email app out there. it is, man. It is. It's more premium, but it does a lot of things really well.

It's a nice experience. But just shifting gears a little bit onto what Arman does with generative AI in terms of like, how do you do this to produce your content and things like that? Is that something you, what kind of tools are you using? Do you have a content workflow these days?

Yeah, I'm I to focus on Steno, like building this startup. I do have a podcast. We have a weekly podcast called Alfalfa. And I co host this with three of my best friends. And every week we talk about money and meaning and we're shifting more, we used to be heavy on the money and crypto and we started in the, pandemic and, crypto is exploding and money was everywhere and people just loved that they want us to talk about that, but the more we had these Deeper philosophical discussions about things, discussions that were difficult to have, especially with friends.

We started to become like these four friends to people. And, I know that's a little cliche, but that's been the audience that we've built. And it's a really hardcore, incredible group of people that are like in love with the community. it's a community first podcast. Like our Discord.

It's unbelievable. So I'm always, I just started because I'm using Midjourney a lot now. I need to get into your podcast. Get in there, alfalfapod.com There's a link to the discord, but it's an incredible community. Even if you don't listen to the podcast about connecting with Really intelligent people who know how to have nuanced, difficult discussions and who are just like incredibly positive and successful.

There's a lot of successful people in there, make good money and have done great things with their careers. And in order to prep. For that podcast, I absolutely, and create content, whether it be for the social media or to show up as a host and have good content that week. Yeah, I definitely have to prep.

However, my tool usage is going to be very boring compared to probably some of the other people you talk to who are really into the AI tools. Because I'm probably mostly building Steno, I'm not as much like in that world of trying so many. Creator tools all the time. My workflow is pretty simple.

I really, like I mentioned, I like Claude a lot. I think for research, I really love Perplexity. I love the co pilot feature there. I pay for all the tools. I always try all the premium for all the tools. And I think Perplexity does an incredible job at helping me prepare with data.

for my content, for my episodes, that I don't just walk in the room like an idiot with just my own opinion. I'd like to go and stress test my opinions with perplexity and see what the research says out there, and then bring that back and challenge myself. I also will go to ChatGPT and challenge myself by having ChatGPT act as somebody that whose opinion I dislike.

And I also have it do the opposite. I like to have that conversation with, the other side of the room because our podcast in particular, we don't lean left. We don't lean right. We are radically moderate. And by that, what we really mean is that there is no black or white. The answer is somewhere in the messy gray, as we say, and we.

don't identify the way that modern society does all the time with Oh, that's my identity. So that's my belief. We think that's very stupid. and we think that's what causes a lot of the problems today. And so we've created a space where we don't really know what we don't know. And we challenge each other's viewpoints in a healthy way.

And even if I have a belief that might lean really left in one, manner, in one topic on the next topic, it's the other way and that's okay. That's just the way it is. And so being able to have that conversation and bring data to it is really fun. So I like to have ChatGPT actwith me and role play with me a lot.

Give me an example of something you, you said you disagree, like something you might not have the same opinion on. Man. Give example. Oh man. If you tune into the pod with me, we're usually almost, I have tuned into your pod a few. I know you have this even, I, saw you like drinking on the podcast too.

Something. Yeah. We drink too. We all, we almost always. Start with we're usually having some kind of scotch or wine or tequila or something like having a chat in the living room, right? Yeah. Yeah. And we recently got like a new studio as well. So it's a cool vibe, we've, we have a drink and.

And, we go from there and we usually have in mind like what the topic is going to be, okay, like the one that we just recorded this week was, are Zoomers and Millennials financially effed? do we have a chance? Do Zoomers actually have a chance to make it financially? In this new world, will they ever be able to buy a home?

Will they ever be able to, rich, truly, with the amount of money that just flies out of your bank account and inflation and living in a big city and making a large salary, but most of it just goes to your cost of living, By the way, San Diego, I'm pretty excited. We got most expensive city in America.

We won, which funny. yeah, not this year. This year, San Diego overtook San Francisco, LA, New York, Honolulu. It's crazy here. And I'm joking. I don't want it to be expensive here, but, we're number one now. And discussing that. And we had a very You know, opposite and opposing viewpoints on like, you know, one person being like, no, stop whining, get to work.

like this generation is weak and that's a common motif. That's a common thing that older generations say about the new generations, right? It's this generation is weak. get to work like our parents did. And then others being like, no, this is completely justified. We are effed. there is no way through this.

Like how, we have literally inflated the economy and The amount of debt that we have, there's no way through, and my co hosts are very smart, and so we have very cool, provoking discussions about stuff like that. Yeah, I will link it up in the description as well. It's pretty funny as well, I've seen, I've watch on YouTube, I think they're on YouTube as well, so I can Yeah, we started doing YouTube as well too.

I like the YouTube platform, but do you do anything to, publish these episodes, you do, you use AI for the prep and things like that. I do that sometimes as well, when I want to dive into deeper to a topic I don't, maybe I'm not, I don't know, or I want to have I like perplexity as well.

I like this actually for getting the data, but for the repurposing or. Marketing the show. Do you use any AI tools or do you have any strategies for that as well to get it out there? Because I know you've been doing pretty well with the show over the years. Yeah. So we've tested a lot of, video clip creation tools.

one example is like video AI. That's a pretty big one. And I actually think that they're getting really good. The challenge is the better your content gets, the more you don't want it to look cookie cutter. and AI is at a point right now where it is getting better and making your content more customizable and unique, where it knows where to find the juicy clip.

It knows how to pull it out and put a little frame around it in captions to make it really cool so you can post to social media. but it's not. Quite there if you want like grade A level content, although I would say maybe in the last like month or two, video has gotten a lot better.

Video has gotten a lot better where you can probably get some like top tier clips made. If you put in the time though, you still have to put in the time and think about it. You can't with AI today, like as much as people want to say Oh my God, it's amazing. And it does everything.

Yes, and if you just take the default thing that it gives you, it's not going to be that great. You still have to do the work. You're the brain here. You still have to be in charge and AI is the tool. And so if you just completely delegate your work to AI, I don't think you're going to get the high quality work that you need today.

To stand out. In the future, different story. It's just like with coding though. if you want to code today, can you actually fully autopilot this and give it up to AI? Absolutely not. If you want to build a new landing page, you could go to, I never know how to pronounce this, Uzard, U U I Yeah, UIzard, U I Z A R D.I O.

it's a great application for web and design. And their AI feature, tested it, went in, tried to build a landing page, and it was I don't know, it was okay. Yeah, I like that you say this, because there are different tools that you can do it, with one click today. I think that's going to get better.

Let's say I have My website, and I want to replicate this landing page exactly. Is AI there to do this right now? No, it's not. It's not going to take like a custom page that I built, like with my designer and things like that, and the same with a video editor, like someone who has been working 10 years plus or something as a video editor is going to be better than Opus Clip or video or something like that to add the transitions or whatever it is.

But I do like where it's at. you are, you have been working as a copywriter for a long time for some big names, so you know. Even if you're charging a lot to your clients, they still need to edit something or hire someone to edit at the end, right? Maybe the last 15 or even 20 someplace.

I had John Benson, he's like a famous copywriter. I spoke to him about this as well. He said, even if they are paying me like 50, 000, 100, 000, I still need to have some editor to go through this. And it's the same with AI. It's like an assistant, right? And if you have an intern in your business, you're not going to give them just one instruction and say, Hey, write a blog post about this topic.

You need to have proper instruction. That's how you give the prompts or the questions or like the details to the AI to actually do a good job. That's right. But I think we're close Navid. I think like being able to. be like, this, these are three badass websites, design, website designs that I like, and I want you to create one for Steno with this copy.

I think we're less than a year away from getting a really good result there. I think that is also going to happen, actually, like you have, maybe you have something you like from other people and you can replicate it. Let's say you're going to LinkedIn and you like a post by, let's say, I like Justin Welch.

I like him as a creator. He's doing really good. And I'm going to ChatGPT or whatever your favorite AI tool is. And I'm asking it to analyze this and breaking it down. So I can understand how he's posting and then create a similar post for me. Let's say I want to write about virtual summits in the same way.

To give some tips in this format, it does actually a good job at that. So I can see it happening for more complicated things as well as creating a website and like doing the same, breaking this down and things like that. I think that's not, too far away, I think. even with just copy, again, your example is great, but if you were able to say, like I, as you said, was a copywriter for a long time, and if I could take all the copy I've ever done and train it to understand my voice and my psychology, because what was unique about my copywriting, it wasn't my writing.

Neil Patel said this a lot when he did interviews and the, testimonials he gave. He said, the reason Armand's copy is the best is because he's the deepest thinking like on the psychology part. Like he finds the unique angle that no one else can find. And that's only because I was just like a psychology nerd.

Like I love cognitive psychology. It's just like a side passion of mine. and I love like human psychology, thinking about how to sell and how to, empathy and how to really dig in with people. It's not about the words, it was about the positioning. And so if you can train AI to get to a point, which I think we'll be able to do soon, to give, here's my corpus of data.

And here's how I approached it and think about it and train it to do that and then be able to provide the feedback. No, you didn't get it. Just again, that's what OpenAI did when they developed this language model, right? It was human reinforced feedback. It would spit out, it's a transformer, right?

It would spit out a result. And then a human would stand there and be like, eh. wrong or, Oh, that's great. more of that type of answer when you're given this type of text. And so it learned how to generate text in a really accurate way. That's why ChatGPT is the best because the amount of training.

And so when you can do that for your stuff and your data or your voice or your design work or your, videos or whatever, you'll soon be able to generate a video that is as good as your incredible video editors, right? Or your copywriters. So that's exciting. No, for sure. I, that's why, usually for my purpose, I like starting with a longer piece like this or something because that can be turned into so many pieces of content.

This is like my main content. It's like unique. It's not like it doesn't exist out there in the way we see it. So I can take that, summarize it or whatever. I can also take, if I wanted to have a long blog post, I can use a tool like Content at Scale, which is like a specialized tool.

Taking the YouTube link, getting like a really in depth blog post from this or a few of the videos I've done. I think that's also becoming, more common, like really specialized tools like that to do a specific thing. Like I haven't, personally, I haven't found any tool that can write.

An in depth blog post based on a YouTube video. And I'm like, all really happy. that would take me forever in ChatGPT to do the exact same thing. It will take, I might get some summaries out of it, but this is like exactly SEO optimized blog posts. And it's it's pretty complex, to do it like the right way.

And not like these long sentences, like Neil Patel would write it or Brian Dean. Like we both know that. that's how you would rank in Google. So I don't understand why some people, they are putting out crap with AI. Essentially. That's something I see a lot because Google doesn't penalize AI generated content.

They penalize bad content, right? That's what they've even said that they were a little bit against AI content in the beginning, I think, but now they're having, they have an open mind to it and they even, they're releasing a lot of AI stuff themselves with Bard and I'm sure a lot more will come in the future.

So yeah, that's what I'm seeing, It's like the same rules still apply as what it did with human written content. If it was bad, it wasn't going to rank well. It was before, like the, like long time ago, probably when, I don't know, maybe when you're working in Google and things like that, people could like rank for, in the early 2000s, people can write a blog post about anything or have an exact match domain.

And they would rank for this. Then Google cracked down on this and they just removed all this stuff and then you had to actually have really good content. Yeah, and it's been a while where you, that's been the game, it's been hard to, it's been hard to game Google other than just having high quality content that follows a few basic principles, of course you want backlinks and all that stuff, but you're going to get those things by having something useful and unique.

And I think it comes down to a lot of the bad quality content that's out there is just because people are not great at prompting yet, it's not natural to people yet. It's going to get there soon, where if the default for you becomes to talk to a chat and prompt it instead of do a search.

then prompting will get a lot better and it'll be a skill that people develop, but it's still early days and there's some good guides out there and I'm sure you'll talk to some people that are very good at explaining how to prompt, and that makes a big difference. I feel also like I've heard some people talk about promptless AI for the reason that it's not like prompts are, it's not, of course you need to have something to give it.

That's a prompt, but it's more like treating the AI as a human and also like in You have a lot of questions. You have a lot of problems built in behind the scenes, right? And I think even ChatGPT might get to that stage to be even more user friendly in the future because it doesn't have a manual, right?

It's not like you're buying like a washing machine and then you have a manual how to use it, right? That's not how it works with. ChatGPT and, but other tools, you might have a set of questions. if I'm going into content at scale, I don't really need to prompt. I can say, I want it in the voice of Brian Dean or like a similar style blog post.

I want this, it just fits out this really incredible post that I can just edit with an editor or something. That's just, it's a tool for replacing writers. So I'm looking for, I'm using ChatGPT a lot, but like I'm using it more for ideation, for like summaries. Things like that, and then I'm taking things to like the specialized tools when I need something really like high, level when I need this for my blog or something that's gonna have leverage for my business.

one way that I use ChatGPT that I'm not sure is very common is I use it as a brainstorming partner and I do have a more natural back and forth conversation with it. In particular, like I don't always love the audio, but have you done the audio? yeah. And so that's amazing when you're on the go, have it on the phone.

I do it. Yes. And so by the audio, not the audio where I can. I don't use my voice, which I use that one all the time because I'm able to then have a very specific long prompt with my voice and think out exactly what I'm trying to ask. But, it was maybe two weeks ago or three weeks ago that ChatGPT launched, Voice back-and-forth where you could go in and pick the voice of the AI that responds to you and yeah, and that's not always convenient, but sometimes like I was like in the car and I just parked my car and I was like doing something I was getting a phone call and I was like, ah, I have this.

I was actually, to be specific, I was trying to calculate the TAM, the Total Addressable Market for one of the sub markets for Steno. And I needed the data and I needed to have a little bit of a back-and-forth conversation about it. And the voice was perfect for that. It was like talking to this friend who's fucking brilliant, excuse me by the way, and being able to have a back and forth discussion about it, in an instant.

Really engaging way and so I think that's another use case that people maybe underutilize sometimes. I love that brainstorming. I use it a lot for like outlines for products and things like that and going back and forth with this. I think that for these things it's really helpful and then I use it sometimes for emails, things like that to get ideas, some clients, all this kind of stuff it's really useful for.

Man, this has been a great session. I just have two more questions. So what's your number one piece of advice for a creator to get started with AI and start implementing this in their business? Number one, I would say, I know that, this is not everyone has the budget, but I would say to try to find as quickly as possible, the tools that are going to work best for your business by signing up for the premium and paying for every tool.

I so often hear people come up to me and be like, eh, I don't really get why this is valuable or that's valuable. It's because they're not invested. In investing, in general, like real investing, we have a principle that if you want to learn about the asset and you want to, yeah, like literally, if you want to learn about Tesla, buy some Tesla stock.

What that does is it's a forcing function to get you to care and pay attention to it. Coinbase, Just buy some, buy 10 of the amount that you think you want to buy in the future, today, immediately. Don't think about it, just do it. Just go to Fidelity, go to Vanguard, buy 10%. And I think it's the same thing for AI tools.

Pay the premium, start the trial, use it. Integrate it into your workflow and see, does this actually have value for me? And that's a way to get to the core of the use case of the product more quickly. Because again, as soon as you've exchanged money, you feel like, Oh, I really have to use this now.

If you just kick around, the free version, it doesn't mean anything to you. It doesn't do anything for you. And so that's probably my number one thing is like pay for the tools. Pay for the tools, use them, and then test them quickly and move on from the ones that don't do it for you.

No, that's a really good advice because also the model is even not even the latest, right? If you're using ChatGPT. Exactly. you're probably spending more money. People complain. I'm not even getting an affiliate commission for it. I just recommend it because it's a great way to get started.

If you use ChatGPT, you get used to it and then you can dabble with other tools as well that are more specialized, I think, even if you are using other tools. Spend the 20 a month. And as a final wrap up question, was there any question that you wanted me to ask, but I didn't, or do you have any final words of wisdom?

Any question I wanted you to ask? I wanted you to ask me about Messi winning the Ballon d'Or. It's been on my mind. Maybe that's another session, man. It's been on my mind. Yeah. You should get into more of your, another session about your full story. I'm sure that will be also interesting to get into.

Yeah. if I'm not building my company, which is pretty much all I do, I'm hanging out with my wife and my baby and I'm thinking and talking about soccer. So I just love. I love soccer and football so much, don't have a lot of time for it anymore. But yeah, Messi winning his 8th Ballon d'Or was a beautiful thing, and apparently a controversial thing, but I don't agree with that.

People wanted Haaland to win, the Norwegian guy. I know, but it was just, yeah, story, fairytale, fairytale year for him. But no, no parting words other than, Thank you if you're still here listening, I appreciate your time and I really do. Anybody who spends time listening and watching content and spending time listening to me, thank you.

If you want, reach out. My email is available, arman@steno.aI That's the best way to get in touch with me and happy to answer any questions and be of help if I can. Thanks so much Arman for coming on and sharing so much value with everyone in this session, I really enjoyed it. And be sure to check out Steno.ai I'll leave a link in the description and anything we talked about, you mentioned a lot of resources, I drop a link as well.

If you have a takeaway or anything you learned from this session, be sure to let us know in the comments there so we can, reply to you and engage with you there. And maybe you have some ideas on how you can monetize your audience in this way based on what we shared in this session here.

And be sure to like the video, subscribe for more if you want more content like this. And I'll see you in the next one. Ciao for now. Bye.