The Koerner Office - Business Ideas and Deep Dives with Chris Koerner - Greg Isenberg on AI Tools That Will Replace You (or Make You Rich) | Ep. #141

Episode Date: March 10, 2025

I sat down with Greg Isenberg to talk about AI, automation, and the tools shaping the future. He started by breaking down his current AI stack - what he’s using, what’s overrated, and what’s actually making him more productive. We talked about the biggest AI business opportunities, how to separate real innovation from hype, and why now is the best time to act on AI trends. Greg also shared his thoughts on the role of design and user experience in AI products and why the way we interact with AI is about to change forever.This episode is packed with valuable insights, don’t miss it.Greg’s Podcast on YouTube → https://www.youtube.com/@GregIsenbergFollow Greg on X → https://x.com/gregisenbergTimestamps below. Enjoy!---Watch this on YouTube instead here: tkopod.co/p-ytAsk me a question on or off the show here: http://tkopod.co/p-askLearn more about me: http://tkopod.co/p-cjkLearn about my company: http://tkopod.co/p-cofFollow me on Twitter here: http://tkopod.co/p-xFree weekly business ideas newsletter: http://tkopod.co/p-nlShare this podcast: http://tkopod.co/p-allScrape small business data: http://tkopod.co/p-os---00:00 Introduction00:35 Greg's AI Stack05:38 The Power of Data in AI08:51 Recording Life with Limitless Pendant11:38 The Future of Note-Taking and AI Tools14:22 Exploring New AI Innovations17:36 The Beauty of User Experience in AI Products21:42 The Allure of AI vs. Automation22:58 Programmatic SEO: A Case Study in Animal Aftercare25:51 The Future of Content Farms and SEO Arbitrage29:12 Navigating Opportunities in a Saturated Market32:01 The Importance of Taking Action Now33:28 Building with AI Tools: The Last Mile Challenge35:14 Innovative Business Ideas and Accreditations40:46 Vibe Revenue: The Reality of AI Business Models

There's an unfair advantage today if you know how to use some of these tools. Are you going to poke holes in that? There's no holes to poke. This is going to happen and someone will make a billion dollars doing this. It's low risk, high probability of success. Putting it in your daily workflow is going to pay the greatest dividends you've probably ever seen. You're required a team to create a product, launch a product. Now, you have people like us sharing how to create products for nothing.

All that's required is motivation. Can I tell you the tool that I've been using? using for the last week that has probably changed the game for me from a productivity standpoint. What is your go to? You have a question. You move your cursor to a tab. I'm sure using all kinds of AIs, but what is your go to? Is it open AI? Is it a deep seek? Cloud, what are we talking? Do I look like a deep seek guy to you, Chris? No. Yeah. No. Thank you. No. As of February 27th, that's when we're recording this. And I say this because I'm constantly changing my stack, my AI stack. And I think that's everyone. I think anyone who's making AI as a part of their workflow,

they're constantly learning like, oh my God, Clough just dropped three, seven. I'm going to go move to that. Or perplexity just launched deep research. I'm going to go use that. So my current stack right now when it comes to, I have a question and I want to figure out an answer to that question. Right now I'm using like deep research for 80% of my questions. And what that means is I'm going to GROC, GROC 3, which is just GROC 3 deep research as recording this, free, amazing. I use chat GAPT Pro deep research when I want like a little extra horsepower. Well, you're paying $200 a month, but I mean you're going to pay in time. It's going to take you GROC3 deep research query takes about a minute or two.

Chat GPT Pro deep research is going to take five to ten minutes. And for like the light stuff, I'm using perplexity still. I really like perplexity. I like the UI. I know I should be using Gemini more. I just don't. No. To me, Google's like the boomer in the room.

Google, we know you have all this data. I don't think you're using it right. I think you drop the ball. It almost feels like they're late. I've been playing with the Google stuff a little bit over the last week. And I will say that it's really good. Notebook L.M. How do you like notebook LM?

It's really good. I think it's like that's their MVP, right? Like so my hot take on on Google is I think that they actually are just, you know, launching the tip of the iceberg. Like I feel like they've, they're probably from a tech perspective, like above and beyond everyone else. You know, their incentive is, I don't know how forward thinking they want to like launch things, right?

Because they have that search business that's making money on ads, right? So I don't know how they want an AI. first world to happen ASAP. But anyways, my point is, for the average person, like, we all should be playing with Gemini because it is really good. How do you like using deep research on perplexity versus deep research on open AI? Deep research on perplexity feels like it's really good. What I'd like about it is you can, it gets the main points across.

and the UI again is just so simple. But sometimes I don't want simple. Like I don't want the headlines. I want to go deep. Like, you know, they call it deep research for a reason, you know. So there's some, I'm not like saying that perplexity is a bad product or perplexity, deep research is a bad product. I'm just saying that when I'm in the mood for like, it's Sunday morning and I need to go deep on something,

I need chat GPT pro deep research. But if I'm like, I've got two minutes before meeting and I really wanted to answer this question, but I don't want like a quick chat GPT or a quick perplexity answer or quick groc answer, that's when I'm going to perplexity deep research. Yeah. I don't know. You probably didn't see this, but about 20 minutes ago I tweeted about deep research first operator on chat GPT.

And like to me, like, Operator came out first. We made some videos. It was fun. To me, it's kind of like a party trick. And deep research is like the replacement because it's, it's just doing all that in the background, but better, but faster without the API limitations without like, all right, you're on Facebook. Monitor the window every five seconds. Like, I haven't touched Operator in like two weeks almost because of deep research. The problem with deep research is the branding of deep research.

Like, no one wakes up every morning and is like. wow, I want deep research. I think a lot of people wake up every morning and they're like, I want agents to go and do my work for me. What people don't realize is that deep research is still agents. There's agents going and scouring the web and putting it in a way and formatting in a way that is easy to consume. So I think if OpenAI launched deep research and called it a research agent, no, we wouldn't

be talking. everyone would be using them. It would be, you know, but I think because it's called deep research, it's just a lot less exciting to the average, Joe. I mean, they nailed the branding with operator and they completely whiffed it with deep research. Yeah, totally.

Now, GROC 3, big step forward, in my opinion, do you think, like, I just don't understand how companies like Twitter are catching up so fast with such a smaller data set. I know they have a lot of data, but it's different data. Like, how is it so good? Well, it's good data. It's really, really good data. I mean, how, you know, Twitter has been around for 20 years almost, almost, you know, closer to 20 than 10, right? I think Twitter came out in 2006 or something. We're talking about billions of tweets and also a lot of new tweets that are super timely. So they've got a really interesting data set. So it's a mix of they've got a really interesting data set and Elon Musk is like Elon Musk.

I heard some crazy stories about him figuring out ways to get a lot of GPUs and creating a, I think they stores them in Tennessee and, you know, making the impossible possible. Elon Musk is, you know, this is not a political statement. It's just like he's able to do these things that a lot of other people can do. I think that's the reason why GROC 3 has become such a good product. I thought GROC 3 was going to suck, honestly. Yeah. It's in the stack.

It's in the daily driver. Is Grock pulling from like externally outside of Twitter as well? Yeah, I think so. Okay. Because I'm over here thinking like there's 80 billion websites and Open AI is building on that. And then there's billions of tweets. If Grock is pulling from a fraction of those websites and then I could see how they could be competitive. You could say that people create websites.

So it's kind of like tweets. But it's people create tweets, you know. I think people are underestimating how valuable Twitter is to creating a high quality LLM. Yeah. What do you feel about this thing, this pendant? Have you looked into these? What is that?

This is the limitless pendant where it records everything. I'm like the total weirdo. I wear it everywhere. I be like, what is that? And I'm like, I'm just like an evangelist because I think it's awesome. And I want to record everything. I don't care about privacy. I'm done like I don't have privacy whatever I want to record my whole life and then like I could write an autobiography with a prompt or I could ask it you know how am I good dad you know based on how I talk to my kids or am I a piece of crap husband was that gas lighting you know like it's just it's just everything I don't understand why more people aren't taken to it is it because of privacy is it because it's just too early what do you think okay walk me through what limitless is and

Yeah. This is not an ad. They don't sponsor me, but I think it's cool. So they're only on test flight right now on the app store. So it is an iPhone app and then a piece of a hardware. It works well, two days battery life. And it's just always listening. You got to keep it within two feet of your mouth. They just came out with an update last week that will identify my voice. So I trained it on my voice, but it doesn't know anyone else's voice. It just says like you and then speaker one, speaker two, speaker three. And it shows summaries of conversations and then you can click into it and open it and it will show verbatim every single word spoken. And then you can ask it questions.

You can say like, give me highlights of my conversation with Nick yesterday or what were the 10 most funny things said in my presence yesterday. You can't ask it to cover like a week or multiple days, only day at a time. So it's still pretty early. But it just records everything. And I just think that's cool. And I don't even know why I think it's cool yet. I just think it's really going to come in handy one day.

I understand why you think it's cool. and I'm a nerd like you. At the same time, I really don't want to wear a device personally. So I don't know if people are watching this on YouTube or on Spotify and Apple, but I have a analog. Of course you do. Cassio, blacked out, you know, because I'm wearing it all blacked out. Wow.

That must have been expensive, man. Dude, you didn't even know. We're talking like... Did you put that on your credit card? I put it on my credit card. and let's just say that I got 100 points. It was 56 bucks, okay? Okay, wow, okay, it's a flex.

It's a flex, right? I have an Apple Watch sitting at home collecting dust. The reason is when I'm in front of a computer, I want to be as plugged in as possible, and I've got tons of apps that I'm constantly using, lately they've been using Granola, the no-taking AI not-taking app. things like that but when I'm away from the computer I don't want to I don't want a device on me yeah

that's that's where I'm at I agree with that my phone my default setting is do not disturb 24 seven I have this watch for running long races but there's no notifications nothing pushes to my watch I'm totally in the camp of like why is it so normal for anyone to get anyone's attention at any moment in the day like that should not be normal or healthy right this at least is not interactive. It's just like a piece of metal, right? There's no touchscreen. It doesn't buzz. It just records. So at least it has that. I think that this is going to be very temporary. Eventually, you know, my watch will just be recording because, you know, if a phone's in the pocket, it's not going to hear. I kind of look at it like I look at biohacking. When I first got into

biohacking, it's like, all right, let's track everything. What's my pulse oxygen level? Like what's this, this is? And then it's like, okay, am I getting fat? No. Do I feel okay? It's just the midwit meme, 80-20, right? So I'm probably just like in the middle of that midwit meme right now. I'm like in my church, Church of Jesus Christ of Latter-day Saints, we're big on genealogy, ancestry, you know, record-keeping. And so it's probably in my DNA to an extent. And my grandkids probably don't care that I had a conversation with Greg today, you know, in a hundred years. I think they will, by the way. I think that Well, because it's you. Well, no, not even.

I think I'm just kidding. I think if I could go and listen to my ancestors or people that have passed, like I would pay any amount of money for that. I'm a nerd like that. Like I need like I'm, I would love to hear that. Here's an analogy. My mom is an avid journaler. She has written in her journal daily her entire life. She has like 150 journals that are this dick.

And I told like every time I see her, I'm like, mom, I got to get your journals. I'm going to scan them. I'm going to transcribe them. And I'm going to turn it into an LLM and I can ask questions. And I, she doesn't really care. She's, you know, it's obviously her journal. She doesn't want it on the internet. But when she dies, and I told her this, like, those journals are all mine. And I just think it would be cool to be like, you know, what were the 10 biggest trials my mom experienced? When was she the saddest? When was she the happiest? And I'm not a journaler. I can't do it. I don't want to. But this is like my cheat code to journaling. You know what I'm saying? So some so my kids can do that to me one day. But I will say,

Greg, let me interrupt myself. I'll take the other side of that argument for a minute. It doesn't get like my insecurities, my feelings, my thoughts. And that's the important stuff, you know, hopefully it can combine with what I'm doing on my screen. It can make assumptions. Yada, yeah, yeah. I hear you. I like, I want the outcome of what you're doing with the limitless, but I don't want to actually wear it is kind of where I'm at. It's the same like someone asked me this morning, oh, Greg, you should totally vlog more because I was telling him a story of how I got invited to this billionaire's house, this famous billionaire's house and all, it was a crazy event. He's moved that way.

And he was like, oh, man, if you would have vlogged that, I'm sure so many people would want to see what it's like to have met that person and gone for dinner and seen some of, you know, the craziness. And I was like, yeah, but then I'd have to like take out my phone or have, like a GoPro type thing and I would feel weird about it. I don't know. It changes the dynamic. There's there's two types of people. There's people like you who want to adopt hardware type devices because you believe that the outcome is worth it. Like the upside is worth it. And then there's people like me who from a hardware perspective like software I'll try anything. Like I will

try anything. But from a hardware perspective, I'm really, if I'm going to purchase a piece of hardware or even if someone wants to give it to me, I need to make sure that the upside of it is absolutely massive. Okay. That's fair. That's fair. That's where I'm at. Yeah. Understood. Are there any like interesting kind of gimmicky AI wrappers or just like these, any interesting tools that you're working with nowadays that are really cool or interesting, either from a user, consumer perspective or from a business perspective. Can I tell you the tool that I've been using for the last week that has probably changed the game for me from a productivity standpoint? Yes, please. There's an app called Whisper Flow. Have you heard about it? It sounds familiar,

but I don't know anything about it. I always used to make fun of the person that would transcribe their voice, you know, go into the notes app and be like, Siri, like, take this note, you know, note for me. My uncle does that. I, it was the, it is by far the most boomer thing of all time. Yeah. I always saw it and I was like, how do you look yourself in the mirror? Honestly. Until I discovered whisper flow, whisper flow, it uses AI to basically transcribe what you, what you're saying. And it does so in a way that is a magical experience. I remember the first time tried chat GPT and I queried it and I got an answer and I was like it was an aha moment. I had the same aha moment with whisperflow. My friend told me to try it and then we started texting back and forth

through voice. I would record it and I would just say like hey like glad you're having a good time on vacation and you know I hope to see you soon and thanks for sending me that picture of my niece or whatever it is. It was the aha moment because it was literally perfect. The commas. were perfect. It took out when I said literally or took out when I said like or and now I'm using it everywhere. I use whisper flow everywhere. When people get slack messages from me, it's likely whisper flow. So thinking about how you use voice as a part of your day to day, I was never thinking about voice. I've introduced voice into my day to day. And people think that, you know, I'm sending these long slack messages that are like really well thought out of. In the past,

would be doing is like one-liners because I didn't have the time to write, right, right, right, in format, format, format, format. With Whisper Flow, and again, I'm not like involved in Whisper Flow. I just think that didn't use voice, now I use voice. We've gone to the place where like Siri isn't good, let's be real. But with LLMs and AI, something like Whisper, Whisper Flow is a game changer. So it basically takes what you can already do in iMessage and it just levels it up everywhere on your iPhone.

Exactly. You can like whisper to it literally. Like you don't even need you know Siri. It's like, okay Siri. You know, hey Siri. And you're got to like literally, oh, my Siri like went off. You literally have to grab a loud speaker and blast your message into it for it to hear. But for with whisper flow, you just quietly say it and it's going to pick up on it. Okay. That's a good one. I'm going to try that. And it's like embedded in your keyboard or it's a standalone app. Exactly. It's cool because it also like it'll like bowl it out. Like let's say I'm sending a Slack message to someone.

Sometimes, you know, I'll say a paragraph or I'll just start talking to it. And it'll just format it in a way that it's easy for the recipient to understand. That's a good one. What else? Anything else you're playing with? I'm trying to think of like non-obvious ones. Or maybe not even ones that you're using, but ones that you've seen in the wild that kind of make you go, huh, oh, that could be an interesting business or that's a stupid idea,

but it has 100K MRR. It's second month. I brought up granola. Have you used granola? I use Fathom. Yeah. Our team uses Fathom. When you go to granola, it's granola.a.i.

I just started using it. Just look at how beautiful the product is. I think the issue with AI products is there's a lot of AI products that create beautiful demos. but the actual product itself isn't beautiful. Granola has been just an extremely beautiful experience. I've only used it for a few days, just that it's beautiful. And I'm excited to see more beautiful AI products.

Another, oh, sorry, go ahead. I was going to say, like, with all these AI note-taking apps, we take all these notes. Everyone's taking notes. Are you ever going back and reading them? Like, honestly? And it's the same principle as this thing I'm wearing. Yeah.

You know, like if I never read this stuff, what's the point? Well, if you're never going to read it, there's no point. There isn't. And that's why, dude, I've been, I'll say it. I'll admit it. I'm not a note taker. Me either. I'm not a not taker.

Never was. Never will be. In school, I always thought it was like, I was a guy to like, there's a girl in the class and I'd be like, you know, next to me and I'd make friends with her or a guy, you know, there's just one girl in particular I'm thinking of that I just went through so many grades with. And she was like my note girl. She took the most meticulous notes. I was like, become friends with the note taker.

Now I don't become friends with the note taker. I'm not sure that I need the notes. If I do need the notes, even if it's once a month or once every three months, it's worth it for that, I think. It's almost like note taking is one of life's coping mechanisms to make us feel good about like not really knowing the material. It's kind of like, you know, I'm going to eat pizza for lunch, but I drank my athletic greens this morning. So I'm good. Right. It's like you're checking a box.

And I use five of them. I'm like, I use a note taker, but I never, ever go back and read them. But it's good to have. It's kind of like a form of insurance, I guess. It's a form. It's schmuck insurance. Yeah. Yeah, that's what it is.

So video is cool, but you know what's better? Long form audio via podcast and my newsletter, tkopod.com. Go there to subscribe for free to my newsletter. It's one email a week, very tactical. And then go to my audio podcast, three episodes a week, stuff like this. You're going to love it. All free. No sleazy sales, bitch.

TKO pod.com. What are you excited about? Well, I want to do. to talk about one more tool that is one more AI tool actually, if that's okay, that I've been obsessed with over the last week or two, which is, have you played with Gumloop? I keep hearing about it from everyone. So there must be something to it. A lot of people. Oh, yes. Okay, yes. Yeah, a lot of people. My chief of staff is using it. Yeah, a lot of people would, will go to Gumloop. Like, a lot of people in the tech world will go to

Gumloop. I think it's Gumloop.com. And they'll be like, I've seen this before. I've seen Zapier. I've seen me. I've seen Lindy AI. Yeah. And you're right. You have like Granola, Gum Loops UI is just a lot more beautiful. So if you're someone like me who really values like beautiful experiences and you're just going to retain better on Gumloop and you're going to end up doing more and you're going to be able to automate more of your business using it. So I've been having fun just like building flows on gum loop and you're talking to a guy who's like used every like played a lot with zapier played a lot with me played a lot with all of these and wasn't the guy who would actually finish them like i would start a lot of things and come up with a lot of ideas and then i would have to go to my team

and be like do you mind just like hooking the stuff up same you know yeah but with gum loop i think what's really cool about it is a lot of people are actually going to finish them start to finish What cool flows do you use with Gumloop today? A lot of scraping data and putting it into Google Docs or Google Sheets. Yeah, just finding data, organizing data, curating data, presenting that data to the right member of our team at the right time. To me, that's the most interesting. What's cool about it is if you scrape data and you put it into, let's say, at Google Sheets, you can use intelligence, you know, in between there, right?

So you can say like, okay, here's all this data that I scrape from, you know, the best Cuban food in Miami. And then you can use intelligence to be like, okay, here's what the internet says, which is basically like highly ranked SEO articles, which by the way, it might not be the best, like, just because it's on a list and it's ranked number one doesn't mean it's actually the best Kubano, for example. And then you can use intelligence and you can. pick which LLM you want to use to actually add to that, edit it, challenge it, put it into some sort of cloud software that, you know, your team could see. And then you can create like Slack bots and stuff like that that notify people. I just think that the concept of when people say they want AI, when business people say they want AI, I think a lot of the times they actually really want automation. But the word AI is just a lot more.

sexy than automation. Zapier has had an open AI integration for like two years now, but it's just clunky. And Zapier breaks all the time. Here's a thought. We started a business, not even a business. We built a programmatic SEO site a year ago for pet cremations. Did I tell you about this already? No.

Yeah, because like there's, there's more puppies than kids nowadays. There's all these animals. There's only a few hundred pet cremation facilities, but the demand for cremation. is going through the roof. And so the thesis was, you know, in all these far-flung cities, there's no, there's no one just starting a horse cremation facility because there's not enough demand, but there's still people searching for it. So we called it animal aftercare. And this was really more of a test, a programmatic SEO learning experience. We built it a year ago. We didn't touch it. And I get calls on my personal phone every hour from people wanting to cremate a thoroughbred

in you name city in the United States or Canada. And it's really interesting because I saw a guy on Twitter doing this with other keywords. And he basically finds keywords on the Google keyword tool that have high search volume, low competition, and then he's using Replit to build a programmatic SEO site. And so what is stopping anyone doing a mad dash, a gold rush, if you will, of taking all those, like high volume, low competition, and high ticket? Orthopedic surgery, for example. and just spinning up hundreds or thousands of programmatic SEO sites.

You do it once in Replit and then you copy, paste over and over. Same design, different keyword, 5,000 landing pages per website and just aggregating countless leads. And then selling the whole thing to private equity, to the biggest player in the market, or just selling the leads. Are you going to poke? There's no holes to poke. This is going to happen and someone will make a billion dollars doing this. This happened in Web 2.4.

2.0 era with, I don't know if you remember, a company called Demand Media. Demand Media basically would use, started in 2006, they would use algorithms to identify topics with high advertising potential based on search engine query data and bids on advertising auctions. So it was basically a lot of, you know, advice and how to the website. So for example, you might have heard of ehow.com started through demand media. I just didn't know it was called the main media. I've heard of these guys. Yes.

So they are credited with spearheading this idea around a content farm. Before them, the idea of a content farm didn't really exist on the internet. The idea that you'd hire writers to write in one of these niches based on Google trends data and keyword data. This is content farm 2.0. I think that there's going to be a demand media of our generation, of this generation. Demand media wasn't a small business by any stretch. I think in its heyday, half a billion dollars a year in revenue. Like we're talking serious revenue.

And this was like pretty year, you know, mid-2000s, right? So that's a lot. It's arbitrage at scale. It's arbitrage at scale. And it's never been easier. Like we will have such an easier time doing this than demand media ever did with these tools. Yeah, I think if I was in college right now, this would be something I would be doing. I'm going to share my screen. This is our Google search console for Animal Aftercare. We had it at

Animal Dash Aftercare.com because that was a $10 domain. You tried it for a month or two. Traffic started going up and then we paid $4,000 for Animal Aftercare.com and started transferring all the traffic. But look at like, I go weeks at a time for getting that I own this business, other than the fact that these calls come in because we don't do anything. We're not selling the leads. We're not doing anything with them. And look at all these keywords that we're ranking for, private cremation for cats, pet cremation, West Covina. Like, because we have 5,000 landing pages in the states, which is five keywords across the top 1,000 cities by population. And then same for Canada.

Top 100 cities in Canada, five keywords. pet cremation, dog cremation, pet euthanasia, dog euthanasia, horse cremation. And we're just ranking. And we just picked a random thing, but you could do this with thousands of keywords. Yeah. What I'd love about this is it's low risk, high probability of success. I mean, the hard part is figuring out, I mean, you've got to use like HRFs or something like that to figure out the opportunity zones.

If you can get the right domain, I actually think that a dot com domain and, you got to and single two-word English dictionary.com domains are going to be more valuable in an AI world. So there's a whole arbitrage there. I mean, we've got like a programmatic SEO agency. We invested in a programmatic SEO agency because we were like, there's going to be a huge opportunity to build businesses on top of these keywords. It's called Borymarketing.com. And I've just been able to see firsthand like the right programmatic approach, how well it works.

And then you marry that with this prompt to software movement that's happening. And it's amazing. So I think that this is a window in time opportunity. I don't know how long it lasts. But for right now, it's definitely a huge idea. So I'll take two sides to that argument. I completely agree. But as far as how long does it last?

Like, you ever seen that viral Reddit post that says it's all priced in? Of course. It's classic, right? So it's like, here's my version of that. we're always early. Everything's early. Right. We're so early. Like we're early on crypto. We're early on e-commerce. We're early on AI. Like you don't know if you're early or late until the history of the human race is played out already. Right. So early is relative to how long that thing will be around. Are we early on laser disks? No. Like it's behind us, right? But like let's let's talk about roll-ups,

right? Rolling up marinas and DSOs, dental offices. And like there's thousands of industries and micro industries and sub industries to roll up that we're going to be rolling up, you know, gutter cleaning businesses in 15 years. They're just because of the sheer number of things there are to roll up, right? And then in 15 years, there's going to be hundreds of new industries that we don't even know about yet, right? So when it comes to opportunities like this, on one hand, I completely agree. If you take like braces near me, like someone's going to build a programmatic site for that if

they haven't already. And they're going to be, they're going to be number one, just like Uber's number one. and have 80% market share and then a bunch of other people will fill in the 20% right so we're probably late for that but there's so many different keywords so many different combinations of keywords that even though we are really for the best maybe the biggest opportunities i think this is still an opportunity in five or 10 years yeah so here's the way i think about it was it easier to launch an iPhone app in 2009, 2010, 2011, right after the app store came out? Yeah, absolutely. It was easier. Does that mean that you can't launch an app in 2025 or 2024, like the Cal AI guys,

and make a million dollars a month or whatever it is they're making? No. Of course, you can still launch a mobile app today and still do really well. The difference between mobile and mobile and this world we live in now is there was back then it required a team to create a product and launch a product and go to market and oftentimes you actually have to raise money to go and do that now you have people like us sharing how to create products or products for nothing and all that's really required is some motivation honestly all that's required is motivation that's literally it's true Yeah. There is just going to be, you know, a lot more competition because there's a lot of motivated people out there. I understand what you're saying and I agree with you that you're early. Even in 2032, you'll be early and there'll be opportunities to build something. But as a whole, it's definitely easier to win today than I think it will be in 10 years from now. I agree. And another thing, though, that I would say is if, you know, we're talking about programmatic SEO and building stuff like that. this on Replit, right? And so in that sense, you don't have to raise money. You don't have to hire a

team. So it's going to get more competitive, more saturated faster, right? But on the other hand, the fact that AI has opened up so many other opportunities that have nothing to do with even websites or programmatic SEO, all the enterprising entrepreneurs out there, we're just chasing so many more rabbits now. The pie is so much bigger that I feel like it's proportional. You know what I'm saying? like even though you don't have to raise money even like it's still it just kind of it's a wash it offsets i want to believe what you're saying is true you know and a past version of me would say don't worry about it you know when the time is right there's still going to be opportunity and i agree but like the reality is there's an unfair advantage today if you know how to use some of these tools

yeah and put stuff out there you could have bought real estate in in Manhattan in 1907 when there's a lot less competition, you can still make money buying real estate in Manhattan today. It's just a lot harder and the price is a lot more. That dot com that you want to buy, like once people realize that buying these dot coms and the keywords are sort of worth it, I'm going to guess at a lot of the prices of dot coms are going to go up. Yeah. Yeah, agreed. And if nothing else, this is kind of intangible, but like you should act now because the more you wait before acting, the more likely you are to never act. So just do it now.

Do it now. That's just bias for action is just best practices regardless. And do it now because you're going to suck for a while. And by the time 2028 rolls around or 2027 rolls around, you'll be good. Yeah. Yep. Yeah. I don't really get the people that are like, yeah, I'll just wait until like the tools get easier.

Or, you know, oh, I use chat GPT and it didn't really work that well. It's like, no, no, no. Like this, all these tools, putting it in your daily workflow is going to pay the greatest dividends you've probably ever seen. Yeah. Agreed. When it comes to the replets of the world, Bolt, WindSurf, what are your go-toes for trying to build an app?

Prompt to code. I think Bolt and Loveable are great places for non-technical people to begin with. Bolt's a client of our design agency, LCA. So I just have a little more on love for that. You know, I went to CS school and I dropped out, but I only had three classes left. So I have like a decent understanding of like technical infrastructure and I don't think about these things. So Cursor to me doesn't feel for a lot of people, I think they'll use it and they'll be like

this is overwhelming. But to me it doesn't feel unknown. I also really like Cursor. What about the last like? I feel like the last mile is just as hard as it's ever been. Like I can whip up something in Replit so fast, but then it's like it still doesn't work. It looks really cool.

Have you found any solutions to that? To me, it's like it does its trick, which is to me I need to get something out and deployed and stuff might be wonky or not working great. But if I can validate that there's demand and I'm like, okay, great, then I can go and hire,

you know, one, two, three developers to take it. The last mile today is still humans, in my opinion. Will that be forever? No. But as of recording this, I do believe that you do need a human to, I mean, it depends on what you want to create. Like, if you're just creating like landing pages or a basic SaaS app, you know, maybe you

can just do it yourself. But once you start getting into a lot of different features and a more heavy software application, having a human in the loop is probably a good idea. I don't know if it's just because I'm an idiot or whatnot, but I have like six projects in Replit right now and I can't close any of them out. I've hired some bounties, but still, even they're not figuring it out. So I do think that there needs to be some sort of like school for this, you know, I feel like there's so many people who have a graveyard of replet ideas or, you know, I feel like there's so many people who have a graveyard of replet ideas or or cursor ideas that are like 90% there.

And I think that I wish there was like a modern day business school that people can like go and go online and I wish it was free. Yeah. And people can like learn to actually like know enough technical know how to take it home. But yeah, I just I think that the tools aren't there just yet. All right. So you got any business ideas for us? I mean, I always had business ideas.

I have a notes file, Apple Notes, where it's just called Ideas, and I've got a million ideas. So let me just pull it up. Any category or just? Anything that excites you today? I'll give this one to you in your audience. I don't know if it's an idea as is an insight, but when I first moved to Miami, when I was walking, and I saw a dental office. And in the window, it said, best dentist in America, 2020 and they had this big plaque.

And I looked at it and then I looked inside the dental office and I was like, oh, it's a clean dental office. It's one of the best, you know, confirmation bias. Yeah, exactly. One of the best dentists in America. I deserve a great dentist, you know. And so I walk in and I was like, are you guys taking customers? Lo and behold, thousands of dollars later, lots of cleanings later, referred probably 25 people

there. I'm a customer and a good customer. Now, is she the best dentist in America? Like, probably not. But it got me thinking that there's a huge opportunity to sell, like, basically sell plaques. My initial idea was selling almost like how Product Hunt has like product of the day, number one. What if you can sell some accreditation to startups? Any niche?

Or you start with one niche, do you think? I was thinking just like startups in general. And it could be as simple as like product of things. the year or product of the year. It's like yeah product of the year. It's like yeah and you can put this like badge on your website and the reality is that's going to increase conversion and in a sea of AI startups like you kind of want to stand out so it's like are you willing to pay $200 so that you could have that on your website and it's going to convert more and also like optimize it around people like doing unboxings of physical award like that's worth. That's where

$200. So, you know, and you know who does this really well is YouTube. They're 100,000. Yeah. Like, did you get one? I did. Yeah, I got one right there. You got one too, right? Yeah, I got one. Yeah. It's nice. See, we're talking about it. I know. Right. All right. I like this idea. And I'm going to, I'm going to hopefully level it up a bit. Okay. Yeah. So, like, AI tool directories out there. You probably see them, right? Different categories, chatbots, whatever. So it could be something like that, a directory, and you could have like a silent auction where I get a cold email.

Let's see, I have a, I have an app that I just release. And I scrape product hunt, this business scrapes product hunt, and they email me, the owner of this new app. They're like, hey, we want to add you to our directory. Congratulations. We nominated you for product of the year. Now this is, it gets a little, gets a little iffy here. Okay.

Let's say there's 20 people on this category of the directory. It's a silent auction. and they bid who wants to be number one. And whoever bids the most is number one, whoever bids second most is number two, and then they get a physical clackmail to them. So it's a total racket. But according to this random website

that no one's heard of, but that doesn't matter. That's just the fine print. They are product of the year. I mean, this is a very easy one to three million dollar a year of business, like super high margins, don't need a big team. I like that it could expand to different categories.

I'm bullish on the trend of like accreditation basically. Yeah, like something in the age of AI, you're saying. Like a human stamp of approval or just kind of going the opposite direction of AI. That's right. I think I always try to think about that around when you're coming up with startup ideas and you see a trend or paradigm shift like AI, you have to be like, okay, what are my favorite AI ideas? And you like list it out by priority. And then you're like, okay, what are my favorite anti-AI ideas? And then you listed out by priority.

So that's how that barbell strategy is something I think a lot about. Yeah, I like it. What about an agent? So you see right there, I've got the YouTube plaque and a Shopify plaque, both smart. An agent that basically links up with Twitter, anytime someone posts a picture or a tweet or something about their Shopify or YouTube plaque, the agent goes and does a deep dive on that. person or that business. How did they get there? When did their Shopify store open? It looks at the web archive. It looks at all these tools that already exist to just do this reverse engineering deep dive of the YouTube channel or the Shopify store or whatever it is and then presents a clean little nice

report. Here's how to copy this business. Or here's how to launch your own Shopify store based on how this random store did it in only 97 days or whatever it is. I like that a lot. And the beauty is you can use tools like gum loop that scrape and and write the articles and write the breakdowns you probably put like a newsletter on top of it too get people to subscribe for like a weekly breakdown or something like that and that's a good one it'd be a paid newsletter where hey we don't know if someone's going to post about this three times a day three times a month but as soon as they do about seven minutes later you're going to have an email in your inbox of exactly how to do that that's a good one What is that company that raised like $50 million for their sales agent?

Just like $10. It's crazy to me. The agents raising all this money. Yeah, it's wild. I'm writing, I'm going to post this newsletter. It's all about this idea called Vibe Revenue. Where can we find that newsletter to read it? Greg Eisenberg.com.

Okay. Vibe revenue. What is that? Just like, we're hopping on the gravy train? Well, you know what it was? I tweeted the other day. I had like too much coffee and I tweeted. I'm usually not this spicy, but I said, fine, I'll say the thing that no one is saying.

There's a bunch of AI companies with $1 to $10 million AAR, raising big VC money on what I'll call curiosity revenue, not real sustainable AAR. Basically, what's happening right now is a lot of people have high willingness to try AI products. They sign up for a free trial, convert, but it's just temporary revenue spikes that don't look like product market fit. And I wrote a whole long post. That's so smart. and it got me thinking about this concept called vibe revenue, which is like people, people have been talking about vibe coding,

which is like using tools like bolt and cursor and replet to like one prompt to software. But I think the other side of it is there's a lot of apps, software companies that are that are being put out there that, frankly, like it's just they're calling it ARR, but it's like people are just window shopping. You know,

they're just trying it out. Yeah. It's like it's a sugar. It's a sugar high. It's a sugar high. And that which is easily gained is easily lost, right? That's right. Exactly.

The harder way is usually the better way. Yeah. So I've been thinking about that a lot. It's really top of mind. Like how do you create an indispensable business and try to like avoid the hype? So I think a lot of people are going to be listening to to this and they listen to, you know, or they're on Twitter or X just seeing other people posting and they're going to be, they're going to feel the FOMO to build AI businesses.

businesses. It's good to feel FOMO in some sense because it increases motivation, which I think is a really positive side effect of FOMO. But it's bad if people end up creating a bunch of businesses, companies that they take other people's money and they end up chasing these sugar highs when the reality is there's no need to chase sugar highs when there's so many indispensable businesses, sustainable businesses that people could be creating today. And if they're listening to this or you on your podcast, they should have no shortage of ideas. I mean, come on. No, I mean, don't give me that I can come up with an idea.

Yeah. I don't want to hear it. Just literally ask Chad GPD for ideas. Yeah. Chris, thank you so much for having me. I know your audience are my people too. I love people interested in ideas and stuff like that. So I'm grateful that you invited me here.

And it's really nice to see you. And I hope you'll have me back on sometime. I would love that. And where can we all find you? On my podcast, Startup Ideas, podcasts. And if you like that stuff, then you'll find me elsewhere. Thank you for coming on. All right.

Thanks, Chris. I'll see you later. All right, what do you think? Please share it with a friend and we'll see you next time on the Kerner Office.