Best of 2024 - Building a Media Empire with Matt Wolfe

You can actually 100% do really well with SEO with AI generated content. Google does not give a crap. And not only does it not give a crap, it has no way of knowing whether or not your content was generated with AI or handwritten by a human. No way of knowing. I know there's AI detectors out there. They suck. They don't work. They're not as accurate as everybody believes they are.

Hey, I want to pull out one of the most popular episodes this year and rerun it for y' all because it was super impactful, not only to you guys, but to me. It's one that I thought a lot about and it's close to my heart because it's also one of my best friends, Matt Wolf. Now, he used to be the co host of this podcast, Hustle and Flowchart.

For a lot of years, and now.

He is deep into the AI news world. He's kind of like the AI news guy, you know, he's the go to Guy. Almost 700,000 subscribers there on YouTube. Millions of people going to his website, Future Tools IO. And in this episode, what we did is we broke down how he's done all this stuff, how he's built out the YouTube channel, his website, how AI and AI content can help you generate but also automate so much in your business. And he literally breaks down all the tactics. You know, how the guy works, and if you don't follow him on YouTube, you know, definitely go do that. But you're going to enjoy this episode. And you, if, before you get in there, I want to shout out the Delphi digital mind clone that you've seen, maybe, or you've heard me talk about. But I want you to test it.

That's the big thing.

So hit pause right now. Go over to your browser and type in hustleandflowchart.com clone C L O N E. That gets you to a page. It's totally free. You can interact text, voice, or video with my AI twin. And it has this digital mind, the second mind, that's trained on every episode, YouTube video and content I've put out. You can interact with it and actually customize it to your situation, your business scenario, whatever you're trying to learn or.

Engage or think about.

Really, this thing will understand you.

Like, I'm a human, you're a human, and it's crazy. You just got to try it out for yourself.

So hustle and flowchart.com clone will get you there.

Enjoy this episode with my good buddy, Matt Wolfe.

Hey, before you get into the episode, I urge you to hit pause and go check out the AI Digital mind. It's like a digital twin of this podcast and a whole bunch of other content I've released. Go to hustle and flowchart.com AI Joe. It's hustle and flowchart.com AI Joe. You can literally chat with this podcast with me and my guests in text, audio, and video for free. If you want one of these made for yourself. If you're Interested, go to hustleandflowchart.com interest and we make these things out. So enjoy.

Let's get to the episode.

Marketers have never been so spread thin. You know, like, with all this stuff happening, technology is going crazy. It's really tough to really focus and get creative and create great content or connect with the customers like we used to or even prospects. So, you know, you're trying to launch a bunch of content campaigns, you're generating leads, you're trying to score the leads. You're nurturing them so much more, but you really don't have enough time to even breathe, let alone do the best marketing out there. But HubSpot is trying to fix that for you. They have a new AI. It's built into the system called Breeze. So it actually combines their marketing hub and the content hub with all these different cool features that help you automate and, you know, remix your content into a whole bunch of different types, score your leads in a better way, and actually have a nice look at the analytics and everything that's going on in your business. So I would recommend to stop spreading yourself out so thin and start making some big moves with HubSpot. You can try it out for yourself. Go visit HubSpot.com marketers to learn more. I want to pause for a moment and shout out another podcast on the HubSpot Podcast Network, your audio destination for business professionals. This one's called OG Mentors. It's hosted by Julie Mayer and Jimmy Edgerton. And these two, they're a pretty cool duo. Where one is, you know, Jimmy is a lot more on the loose side. I would say that Julie brings it in and keeps us focused on mentoring and movement. And so these folks, they. They focus on mentoring and exploring ways on how to augment and refresh, kind of like the way that mentoring is applied. And they have these principles that they weave into their episodes. They. They give a bunch of practical tips and they lace it with a whole bunch of humor as well. It's always cool when there's a cool duo happening there. So if you're in the mentoring space or if you have a, you know, leadership space, relationship guidance and things like that. That's what the podcast is made for. So you'll get a lot for it. It's great for entrepreneurs, marketers, sales professionals. They love the show as well. It's a great one. I've checked it out and I recommend it. So go listen to OG mentors wherever you get your podcasts.

Mr. Lobo, Mr. Matt Wolf. How are you, sir?

Mr. Fierro, or as I call you, Jose Fierro.

That's it. Ah, feels like home. Yes, back at it. It's good to have you. And of course, we are probably gonna. I don't know how long this episode will go. We do have a hard stop, which is probably good for everybody here. Probably. So we've already been chatting for like 40 minutes without recording.

Yeah. If, if, if, if we could, we'd probably do like Joe Rogan style four hour episodes.

Yes, yes. Well, that's what, that's what we were doing. Well, we didn't, we never did that. I think our longest was like two and some change back on this show.

I feel like we might have done like one or two three hour episodes, but we broke it up into like two parts.

If we did, everybody thanks us for that.

Yeah.

So you're back for those. The. The minority or maybe majority who knows that, that know or don't know you? This is Matt Wolf. He was my co host for many a years starting off the Hustle and Flowchart podcast. And you know, we've been partners on various things over the last, I don't know, 15 years or whatever it's been. I don't know, I stopped counting. I'm like, this show, I was looking at the sign over here, you can't see it. I was like, 2017, because I've been saying like seven years, eight years. I'm like, no, it's nine years just for this podcast.

Like, yeah, that's crazy. That's crazy.

Yeah.

I mean, I know both of us sort of quit our day jobs, so to speak, in 2009. So we've both been full time, let's just say, entrepreneurs for the last 15 years.

There it is. So, yeah, it's legit. It's crazy. So brother from another mother. We've always shared a lot of thoughts, different types of brains collaborate, and that's what's always been fun to jam. And that's why it's like, doesn't feel like we skip a beat, even though we're doing different stuff. Yeah, you're. You've, you know, obviously done some awesome things on YouTube. A lot of people follow you for the AI stuff, as they should. And if you're not, look up Matt Wolf, W O L F E on pod, on YouTube. Future tools. I mean, it's referenced all the time, you know, for AI kind of database finding what you need. Cool tools out there. I just feel like you've, like, you found the thing that you wanted to do for so long.

Yeah, no, I think. I think that's true. Like, I've always been a sort of tech nerd, right? Like, probably as long as you've known me, I've always been like, early adopter. Get all the gadgets, play with all the toys, test all the software. Right? Like, that was. I feel like in the early days of creating content, people always came to me for, like, what WordPress plugin should I use for this? What tool should I use for this? What's the best? Like, funnel builders. And people kind of always looked at me as the guy that just played with all the toys and, like, knew which toys did what. Right. But at the same time, I was always really into content, right? We had the Hustle and Flowchart podcast. I had YouTube. My first YouTube video I ever published was back in 2009. You know, we were doing blog content together. So I've always loved content and I've always loved tech and gadgets. And I feel like when AI started to bubble up, you and I kind of talked about AI, like, really early, right? Back in 2021 or so, we were talking about things like Jasper, which even before that, when it was conversion AI and then Jarvis and then Jasper, right? So we were talking about AI, I feel like, way early before anybody else was. And then I really started to get into AI because of, like, all the sort of visual stuff, stable diffusion, mid journey, like all of the cool, like, AI art stuff. And when all the AI stuff started to bubble up in the mainstream, I kind of found that thing that was like, all right, I get to make content about the stuff that I'm really enjoying now. And I found that sort of like, Venn diagram overlap of like, this is. This is my jam now. And I think it. I don't know people, for whatever reason, I still don't understand it to this day, but for whatever reason, people gravitated to it and enjoy the content.

What do you think it is? Like, have you figured out why you.

No, I ask that question constantly. Like, I have no idea why people Enjoy my content. Like, I actually, when I'm making my videos, I, in my head I'm usually thinking like, why would anybody watch this? I don't, like, I, I don't actually think my content is good. So I'm always surprised that people watch it and leave good comments. But I don't know is that if that's just like I'm too close to my content, so I don't really see. A lot of times I feel like I'm talking about stuff that to me is just like, well, I like everybody already knows this. Like, why am I even making a video about it? Everybody already knows it. And then I put the video out, everybody's like, oh my God, mind blown. I didn't even know that existed. And so I think a lot of it is.

I can speak based on the feedback I've gotten from other people who watch my channel.

Based on the feedback. People seem to like the sort of excitement, the like sort of little kid on Christmas opening a new present energy. Like, oh, look at this new thing. This is awesome. Like, how do I, how are people not talking about this? Like that kind of energy combined with just totally keeping my finger on the pulse probably to a level that most people aren't willing to do well, that's what I think.

That's a good feedback that they're giving. And I think it's honest too. Like you're not boring. Like if it was just you talking head, no editing, you know, maybe screen capture tutorial, like you would still do pretty well I think. You know, because you, you've always been great at that. But then with the quick cuts, with the B roll, with all these other like little silly thing, you know, things that'll pop up or sound effects like, yeah, okay, now you just took it up that extra, I don't even call it like extra 20%. I feel like it's way more for YouTube, like engagement wise.

Well, you know, you and I, we've both been in the digital marketing world for 15 plus years at this point, right? So it's like we learned all the little like marketing tricks to grab attention. Like, you know, when I, when I'm thinking of titles for my YouTube videos, it's exactly the same as thinking of subject lines for emails. Like, what is the thing that's going to get somebody to click this email and open it? Well, what is the thing that's going to get somebody to click this YouTube video to watch it, right? So like all the sort of copywriting and intrigue and persuasion and all that kind of stuff that we learned is all the same for digital marketing that we did for 15 years that I'm doing in YouTube. Right. When it comes to creating video content, it's all the same strategies we've learned over the last 15 years of like, making sure there's pattern interrupts, how do you hold attention, how do you, you know, how do I make sure that I'm teasing out what's going to come later in the video, how do I open loops? It's like all the same stuff that we use when writing copy for a sales letter can be sort of applied over to YouTube videos. Pattern interrupts, hold the tension, open loops, all that kind of stuff. And so I'm constantly thinking about that stuff when I'm making YouTube videos. I'm thinking of it as like, okay, I've got this sort of pool of information that I want to share in this video. How do I make it, A, entertaining and B, make people want to keep watching it? How do I make it so that, you know, three minutes in they're not already, like, glossing over and bored of it? Like, that's where the pattern interrupts come into play. That's where the open loops come into play. Right. So I'm just applying all the marketing skills that we've learned and all the content creation skills that we've learned and just mashing them all together around the topic of AI.

Yeah, and you've done that well. And I know we. The last episode we recorded here, I don't know, six months maybe, plus it's been a little bit. But we talked about that, like, how it was like the three parts of making a great YouTube video. It's basically title, thumbnail hook, Right? Anything changed on that?

It's pretty much exactly the same. I mean, title, thumbnail hook. In fact, I even think that focusing on retention is less important than it used to be.

Retention in the video?

In the video. Yeah, I think it's less important than it used to be because I think people have gotten overwhelmed and sort of they're just over what we call retention. Editing and retention. Editing is essentially like quick cuts, fast motion move. Like Mr. Beast has been like. People have used the term the beastification of YouTube.

Right.

Beastification of YouTube is like editing for retention, where it's all quick cuts, fast yelling at the camera, lots of B roll, lots of text on the screen, you know, explosions, like just so much happening at once where your brain is just like overloaded, like, oh my God. Yeah, there's a lot going on. I Can't look away, right? And I think people are kind of over that now. And there's been this shift back to, like, what YouTube was like 10 years ago, where it was more like personal, right? The Casey Neistatz, the David Dobrik, the, like, the vlogger style where people are just sort of sharing insights into their daily life. You look at people like moist, Critical, or Penguin Z0, right? His videos are just like a dude talking to the camera and commenting on, like, whatever the latest news is. He's one of the most popular YouTubers in the world right now. And he's literally just like, face to camera commenting on stuff. No retention editing whatsoever. And he's one of the most popular YouTubers. There's another YouTuber named Sam Solick right now who's in the fitness space, and his videos are often an hour long, and it's him just talking to the camera and, like, taking his camera to the gym and showing his workouts and stuff. And so some of the most popular YouTubers right now have started moving away from that retention editing and sort of building a bond with the audience. Like, when you're not trying to. When you're trying to do retention editing, you don't really build a bond. You're just going and making these videos that are fast and keep moving and try to, like, overstimulate the brain. So people want to keep watching, but people don't necessarily fill a bond with the creator. When you sort of get rid of that retention editing and you just focus on being a real person, focus on being that dude that other people want to go and sit down and have a beer with, right? That sort of content is what's rising to the top at the moment. So I've been kind of leaning into that, and I feel like that's sort of my natural way of creating content, is just like, I'm going to, like, you know me, right? The way I talk when I'm being recorded is exactly the same way I talk when I'm not being recorded.

Yeah, you are really good at that. And you have been for a long time. Like, you just have that excitement. It legitimately is that kind of giddiness and that comes through. And like, for me, I've had to work on it over the years where I'm like, a lot more subdued, but then it's like, no, hold on, you know, and. But now it feels natural. How have you changed, like, to be, I guess, your videos or like, to have more of that bond? Like, what shifts are you making so.

A lot less cuts, right? So I. I think a lot of people. When I first started making AI videos, I think a lot of people thought I was trying to do retention editing, but that was never really like, my focus. I wasn't trying to edit to, like, keep, like, viewers attention. My edits were because I just cut a lot of stuff out. So if you watch one of my YouTube videos, it might be anywhere. I mean, I think I try to average the videos around 20 minutes. Sometimes they'll be a little longer, sometimes they'll be a little shorter. But the. If you saw the RAW video before I did any editing, it's probably an hour and a half long, right? I'll flip on the camera and I'll talk for an hour and a half, and then I'll go in and I'll just like, all right, that's fluff. That's just me rambling. That's me. I already said that early in the video, I can cut that, because that's just redundant. Right. And so I'm cutting all that stuff out, and when I cut it all out, it starts to look like it's choppier and sort of edited for retention. But that's really me just, like, trying to cut out the fluff. Well, I think what's changed is I've gotten better at just not putting the fluff in in the first place, so I don't have to cut it as often. Right. I think. I think it's just been a. More of a. Like a practiced thing where I don't necessarily need to cut as much because my brain is getting better at getting to the point, not duplicating itself, not repeating myself, and just saying what needs to be said and moving on. And so it starts to feel like I'm doing less and less of the retention editing, but it's just. I'm getting better at speaking on camera.

There you go. Yeah, it's definitely practice. I came across here with. With podcast stuff. I know this has been practiced for both of us over the years doing this show, and. And I know it's made me improve as a speaker and. And just I think confidence, right? Like, you just catch these ums or us or filler words that naturally you might say. And I definitely still catch them.

Yeah.

Power of editing.

You still need, you know, editing. AI editing especially, is quite easy. And these days, just kind of thinking, okay, so. And I love the I think of lifestyle. It's like, it's bringing the human connection. Like you mentioned, bond and community. That's the Case if, like, for course, creators, for instance, actually had a Marisa Murgatroyd back on here recently. And, you know, it was five years before that. And, and even Sean Cannell, he was saying he's like the big thing with courses in anything you're selling digital product. Like, you need to have community.

Yeah.

And. And goes for media as well. Like, how do you. So you have your email list, but it's like, talk about the. The community side and, And I kind of have some ideas, but I want to hear how you're doing it. Yeah.

So I. I put a big focus early on when I started the YouTube channel on Discord as well. But I think. So Discord and X are probably where I focus on community the most. If I'm being totally honest, I don't pay as much attention to the YouTube comments as I used to when it comes to YouTube comments. What I've sort of learned is that very similar to people who leave reviews on restaurants, they'll typically only leave a review if they have a bad experience, not when they have a good experience. Right. So, like, a lot of times you'll see reviews and you got to take them with a grain of salt because most people that didn't like their experience are the ones that are most likely to leave a review. People that liked their experience, they're just. They just go on with their life and, you know, don't feel the need to go talk about it. Right. YouTube comments are very, very similar. People tend to only leave comments if they disagree with something you said. Right. So most of the time, I'll post a YouTube video, and the comments are 90% the people that had something to disagree with something I said in the video. But that's like 3% of the people that actually view the video that actually leave the comments.

Right.

The other 97% enjoyed the video and went on with their life. So for that reason, I actually don't really pay a whole lot of attention to the YouTube comments anymore, saying that I do have a team member that actually reviews them. And if somebody's like, leaving good, valuable feedback that is, like, could help me improve the videos, they're pointing me to it. If people are saying good things about the video, like, they're. They're helping me with that as well. There's also AI tools out there now that can do, like, sentiment analysis on your YouTube comments. But most of the community building that I do is on X and inside of Discord. And the way I do it is I just try to make myself as approachable as possible. Like, if people like at me on Twitter or X, I pretty much reply to everybody. I'm reading the comments, I'm replying. Same with Discord. If somebody's asking me a question in Discord, I get in there, I'll answer the questions. I'm not reading the entire sort of feed that's going in discord. We've got 15,000 members in there now. So, like, the feed is just too hard for me to keep up with. But I do have team members that are like, hey, this person just asked a question. You should jump in and answer it real quick. And I will. And so I think the community element of it is really just me saying, like, hey, look, I'm a real dude. I'm here. Come ask me questions. Come talk to me. Like, let's interact. I go to a lot of events, right. I was at Augmented World Expo the week before recording this right now. So the week before that I was at a Cisco event in Vegas. The week before that I was in Seattle for a Microsoft event. Before that I was in San francisco@google IO. I was at.

Tonight you're going to something.

I was at Nvidia gtc. I was at, like, I've, like, we're. We're recording this in June right now, and I've already been to about 10 events this year, so I'm actually out there shaking hands, kissing babies, as our old friend, fun uncle would say.

Yep, yep.

I'm like, I'm meeting people in person and going and having dinner with them and, you know, doing what I can to hang out. Like, I'm. I try to be as approachable as possible. And I think most people are often shocked that, like, oh, when I hang out with him in person, he's like the exact energy, the exact person that I see in those videos. And that's. That's kind of what I'm going for. And I don't. I wouldn't say it's methodical. I think it's just who I am.

Yeah, yeah. And that's why this whole model is for you. And that's. And, like, that's why I tell people as well, you know, if they're, hey, how's Matt doing? Like, he's literally doing what he wanted to do for a long time. You know, like, it's perfect and.

Yeah, yeah, it is interesting in that way. Right. Like, I feel like at Evergreen Profits, we were constantly talking about the idea of, like, building a media brand, going in harder on YouTube and creating more video content and all of that kind of stuff, and I just kind of leaned into that, and it's. It's been working out pretty well.

Well, and you also carved out a very, you know, a niche space. Obviously, it's kind of broad, but, like, in the time, like, there was a gap, right? Yeah. On YouTube, and you filled that. And, like, this is what kind of what we were talking about before recording is the fact that, like, if you could spot something like that and you can move quick, being, like, as a single person going after it, or maybe have a small team, maybe not a bit. Like, you could see maybe some of the trends at bigger companies are doing, but you're like, oh, wait, hold on. No one's really attacking it in this way or this. This vertical for sure on YouTube, you know, and like, twice a week or whatever it might be like, I feel like that's where you, like, you found the gap and then you just freaking hit it. Yeah. Yeah.

I mean, I don't. I don't know if it was like, that methodical at the time. Right. Like, I was just kind of like, hey, this stuff's really, really cool. I like making videos about it. There was maybe like, one or two other people on YouTube that were talking about AI, but it wasn't the main focus of their channel. It was just kind of like, from time to time, they'd post a video about AI And I kind of decided, I love this stuff. Let's just kind of make the channel all about this, you know? And I don't know, I might have been one of the first, if not the first channel, to really say, all right, I'm going to put my channel fully focused on AI. But it wasn't like a methodical thing. It was just like, that was what really interested me. And I started making videos about it. The video started doing well, and I went, okay, well, these videos are doing well. Let's keep making videos about it.

It's cool. Because Stepping stones. I mean, I don't know who said it, but it's like you can connect the dots, you know, going in reverse, and it makes a lot of sense.

Yeah.

It's never a straight line or whatever. It's that whole zigzag thing. And I mean, we haven't mentioned. But you have. You know, you have a podcast as well, the Next Wave, and that's with HubSpot, which is. Which is rad. Yeah. And that gets a lot of attention. You have the YouTube channel. We already mentioned the community. You have what Future Tools.

Yeah.

Is. Is the website you Also have an email newsletter. And it's like you start. Well, I was mentioning it to you and I don't know, I don't think we chatted about it before this, but I see it as like three key components. Is like, you have the media side, which is your YouTube channel mainly, and then you have your website, which is the hub. You can have sponsorships, other monetization. You can obviously on YouTube as well. And then, and then the email newsletter is all tied together again, sponsorships, but they're all very scalable. A couple of those are sellable. Maybe the media side too, but like the other ones. Yeah, but like that whole trifecta, I feel like, is a model that a lot of people and something that we talked about a long time mentioned, Evergreen Profits, like our old company. It's kind of the same idea, but you said it wasn't really super methodical, but you did it and like, you did it anyway. It's probably in there somewhere, you know.

Yeah, yeah. I mean, I, I, A lot of times I don't think we give ourselves enough credit for, for, you know, how, how it all came to be. Right. Like I say, it's not super methodical. But Also I did 13 years of marketing, content marketing, copywriting, all of that kind of stuff. So it's like in my brain, the pieces all kind of fit together and made sense of, like, what I should do. But it wasn't like, okay, I sat down with a pen and paper and said, all right, I'm gonna make this three pronged approach of YouTube website, newsletter. It was just like the marketing brain in me who's been doing this for so long knew that, okay, we've got the media outlet, which can drive attention. We've got the newsletter, which can drive retention. Right. Keep bringing people back to whatever I'm doing over and over again. And then we've got the website, which can sort of capitalize on SEO and other traffic methods to bring people in as well. But also the website is what grows the newsletter the most. Most people think, oh, the newsletter probably grows from YouTube. I get almost no subscribers to my newsletter, even though it's my main call to action in my YouTube videos. Most of my newsletter growth comes from the Future Tools website. The Future Tools website gets close to a million visitors a month straight up from Google SEO, Right. So it's like I have a way of capitalizing on the SEO with, with the website. It grows the newsletter, which helps with retention, so I can bring people to, back to YouTube videos and back to the Website. And it's just this, like, symbiotic process that to me, just all kind of made sense. Like, if you have a website, you should have a newsletter. We've, like, we've preached that for over a decade now. If you have a website, grow a damn newsletter that helps you bring people back to the website. Right. Content marketing is, in my opinion, the best form of marketing because it's free. You put the content out there, it could live on forever and continue to drive traffic forever. We've preached that forever. Right. So it's like all of the stuff I'm doing is literally the stuff that you and I preached on the Hustle and Flowchart for, you know, however long. We did the Hustle and Flowchart show together.

Nine years with you, about seven and a half or so. Yeah, yeah.

So it's all. It's all just doing, like, practicing what we've been preaching. Right. You build media, create content, grow a list, and that's really it.

Well, so talk about. And this is great. I love the fact that SEO, because.

I was unaware, I kind of had.

A feeling, you know, in terms of traffic SEO, because of the sheer amount of tools that you're referencing daily, like new, new ones all the time. And you obviously have a system for this. This is where you always had your brilliance. So I guess walk through, because a question a lot of people have, and I've heard it a lot asked me, because still, the main driver actually of growth of this show is on the website. It's through SEO as well. Same thing. But obviously you. You're capitalizing like other brands, which is always brilliant. Again, what this podcast has done, too, leveraging on course, break down the whole SEO thing. And like, so obviously it works still, because that's the biggest question. People are like, is it even worth it? Yes.

Yes. Well, it is for now. I don't know. Long term, it's going to be worth it. But right now, in this current moment in time, it's still worth it. So, yeah, like the SEO. So there's been a lot of talk about how like, AI is. You can't create content with AI and get it SEO'd. Right. A lot of people have claimed that it's BS. You can actually 100% do really well with SEO with AI generated content. Google does not give a crap. And not only does it not give a crap, it has no way of knowing whether or not your content was generated with AI or handwritten by a human. No way of knowing. I know there's AI Detectors out there, they suck. They don't work. They're not as accurate as everybody believes they are. There's been stories out there of college professors failing everybody in their entire class because he thought they wrote it with AI. Because one of these AI detectors claimed all the fricking articles were written by AI, Right? So, like, the. Those detectors just. They don't work. If you wrote something with AI and you change like two words in the article, all of a sudden the detectors can no longer tell that it was AI written. So AI 100% can still work.

Or SEO, what are you using? Specific tool, obviously have a whole automated thing, but to do your writing, like, do you prefer any one tool?

Yeah. So I built a workflow. Are you familiar with make.com yeah. Yeah. So I built it for anybody who's not. It's very similar to Zapier, if you're. If you're familiar with Zapier. Make.com is a competitor. They kind of do the same types of stuff. So I built a workflow in make.com where whenever I come across a new tool, I take the URL of that tool, I put it into a Google sheet, into like a cell in a Google sheet, and then make.com takes that URL, uses a tool called Scraping Bee. It uses the API from Scraping Bee to scrape the sales page. It basically just looks at the URL and goes, all right, we're going to take all of the content that's on this page. It takes all of that content, and then it pulls it into GPT4. And then there's a prompt that it automatically submits that says, summarize what this tool does down into one paragraph, and then it summarizes it down into one paragraph, and then it takes that output, that one paragraph, sends it to another GPT4, and asks it to summarize that one paragraph down into one sentence. Right? So now all I do is I plug in this URL into a Google Sheet and it writes me. It scrapes the site, writes a summary, writes a single sentence, and then the final step on make.com is then it then adds it to my website automatically for me. So it takes that description, it takes that short description. That's what you see on the main homepage, is the short description, and it builds out that page for every single tool.

That's dope.

I love that. So that's the workflow you just throw it into. Whenever I find a cool tool, throw it into Google Sheets, and it's on the website.

Wow. Yes.

And so that's sort of how that works. And all of these things rank. And a lot of times if you search for a specific tool, the number one listing is like the tool itself. The number two listing is the future tools listing for that tool. Right? So I know it really works. And what Google really wants to see is if somebody clicks on a link and they go to the site, how quickly do they come back to Google after they click on the site? Right? Like that's pretty much 90% of SEO these days. You know, people who sell SEO services don't want to admit that that's what it is because otherwise everybody can just do it themselves. But that's what SEO is these days. If I find a resource on Google and I click on it and then I land on a page, if I click right back to Google within five seconds of landing on the page, Google thinks, okay, we showed them a result that wasn't relevant to their search. Let's push it down the ranks a little bit. Right? And this is, this very oversimplified. There's more nuances to it. But for the most part that's a very simplified version of SEO. If they click on a link and then they bounce back to Google real quick, Google sort of deranks that site a little bit. That wasn't the right result. That's all Google cares about. It doesn't matter if it was AI written. It doesn't matter if it wasn't AI written. However, if you spend a lot of time doing like AI written blog posts and these days people are getting better and better at going, this feels like it was written by ChatGPT and they bounce off very quickly. Those sites derank. So if you're using AI to write a blog post and people land on it and they start to get this impression, this feels like it was written by AI, they're going to bounce. That site's going to lose rank. And you know, Google's going to rank the higher value content that's on there. Which is why a lot of people say, like, AI content doesn't really work for SEO. It's just because sometimes AI is writing really crappy content that people bounce away from.

And I would, I would urge to say that maybe people either a, they don't know or they're being lazy and they're not humanizing the content, they're not writing it, training it in your, your writing style, or at least prompting it in a way to model someone else. That doesn't sound like, you know, maybe it's using delve five times inside of an article. It's like, when the hell have you used delve in your life?

Yeah, yeah, yeah.

There's certain words AI loves that everyone's kind of picking up and seeing now.

Exactly. Yeah.

Okay. This is great, man. And that whole workflow, I mean, someone could just take that right there and like create a whole workflow of yourself, like whatever application to your website. I know, make.com. it's super simple. It's all like, no code, drag and drop. Yeah, yeah. Is that what your website's made on as well is make or.

No, it's built on webflow right now. But saying that I'm actually working with a developer right now and we're actually going to migrate the whole thing over to WordPress so it's actually going to be on WordPress again, basically because Webflow has like bandwidth limits and my site gets so much traffic that I'm constantly hitting those bandwidth limits.

Good problem? Yeah, it's a good problem. Great. Beautiful, man. And that's the thing, because I think you've had a workflow kind of like this for a while on your site, I believe. And I mean, honestly, it's like any databasing in general.

It'S just clever. I mean, I'm already getting a bunch of ideas down the rabbit hole now, but beautiful. Okay, thanks for clearing up the SEO thing on the website side of stuff.

Funny thing is, I'll just add this too. Is like, I built a website like that back when we were still doing Hustle and Flowchart, but I was having it show pages like lead page tools like LeadPages and Click Funnels. And it had like a filtering thing where, you know, are you looking for a landing page builder? Are you looking for a checkout cart builder? And we'd try to push him to ThriveCart. And I built like the same sort of filtering tool. I don't remember what we called it at the time.

Oh, shoot, it's coming back to me. It was some affiliate.

It was an affiliate marketing play where like you filtered down to the exact tool that you needed. And when it found the tool that you needed, you clicked on it and it was our affiliate link. Right. And so when I built future tools, I literally just took that same concept. I'm like, man, I can build that same same thing that we were building there, but let's just focus on AI tools. Same exact concept. I just sort of applied it to a new sort of niche.

I love it. It just, I. I love the fact that because we proved it on the podcast side as well. And this is through a little bit more manual work because you have to record the content, of course. But, you know, like leveraging people's brands and names in YouTube, like, that's such a powerful thing because we would specifically title episodes. And obviously in the content itself, you could do some more SEO engineering, the fact that you don't even have that much content inside that website. But obviously it's very, like you said, time on site. So they're going to start clicking around other tools and going deeper. And like, that's the point, I think.

Yeah, and that's. And I think that's what happens too. I think people look at, they. They find the tool on Google, they click into it, and then when you look at any tool on future tools, and bottom of the tool page it says other similar tools, and people go, oh, like, like, I was looking for a tool that does this. Well, here's two other tools that do the same thing. Maybe one of these is a better fit. And so they start diving deeper into the website and they don't back bounce back to Google very quickly. So the site stays ranked.

Well, yeah. Okay, cool. All right, so I want to move away from the website stuff now. This is great. No, this is beautiful. Like, we can keep going down, but I'm going to stop there on the, let's see, YouTube side, I think we. We nailed some good stuff there already. There's obviously changes now. You know, like, you put multiple thumbnails in split test those. So there's ways to kind of optimize the whole thing.

I got that feature really early. I've had that for about a year now. And they just rolled it out to all of YouTube within the last, like two weeks or something like that.

Yeah, yeah.

But, yeah, I've been doing the split testing for a long time now. And yeah, it's been awesome. You learn a lot. And what you find out by doing a lot of YouTube split testing is that almost never is the thumbnail you think gonna win, the one that actually wins.

Are there any, like, I don't know, best practices, if that's not the right word. But, like, have there been some commonalities that are the winners or is it literally a crapshoot?

It feels like a crapshoot at this point, I would say as far as best practices go, one thing that we learned pretty quickly is that when you first start doing a thumbnail test, test three dramatically different thumbnails. Right. Because in the beginning, what we were doing is we were doing like three of the same thumbnail. But we'd try like different words, right? Like I might do one that says AI news, one that says crazy AI news, one that says huge AI news. But other than just the text, the thumbnail was identical in all three. Right. We'd go and look in and it would be like 33%. 33%, 33%, right. They'd all be like the same click through slash watch time. It's not based on click through, it's based on watch time. But watch time is factored into click through rate. Because they wouldn't have ever watched if they didn't click through in the first place. Right? So but we would find that they would all be like equally matched. But then if we would put in three dramatically different thumbnails, like, one is like my real face, one is an AI generated face, one is just like a colorful picture and it doesn't have me in the thumbnail at all. But they'd be wildly different. We'd find you'd get like, one would be 25%, one would be 45%, one would be like 12% or whatever. I don't know, I obviously didn't do the math in my head right there. But you know, that one would be like a dramatic winner. And then once you find that dramatic winner, you go and take the thumbnail that won and do another split test. But this time try testing like different text. And now, now you're going for marginal improvements, but you're not really going to see the like, marginal improvements by 3, like really close thumbnails.

So that's why, starting with the variations. Yeah, yeah, absolutely. So one thing, and I think you, if you have a YouTube video on this, just say, I have a YouTube video on this and I can find it. But like, you've gotten, I think you're the best, or at least one of the best people who can, you know, get your face in AI form. But like do some wacky stuff, modeling different characters, whatever it is. Could you break down your process if it's simple enough to do that, or at least point us in the direction.

Yeah. So I do have a couple YouTube videos about this, but the, the sort of quick overview of it is you have to use a tool. So are you familiar with stable diffusion?

Yeah, I use it a ton.

But yeah, yeah, the AI art generation tool. So the way I do it is I have stable diffusion installed locally on my computer. Now if you're using like a Mac, it doesn't work very well because it like they want to see an Nvidia GPU for some reason.

Right.

But I have it installed locally. And what I essentially did was I fine tuned a stable diffusion model and trained my own face into that model. And then once my own face was into that model, I can now prompt any image, use a little keyword that tells it that I want my face in this image, and it will put my face into that image. So, you know, it's probably too complex. Like, I did an interview on Social Media examiner with Michael Stelzner, and that entire interview was like a step by step breakdown of how I do this. And that was the entire interview. Right. So that was like an hour long. Like, here's how to do that process. And then I also have a couple YouTube videos where I break down the process as well. But essentially I'm fine tuning a stable diffusion model with my face trained into it. Once you have it trained, you can literally type any prompt you want and put your face into that image.

Okay.

Yeah.

Well, I just found six days ago, at least time of this recording, repurposing video content into multiple mediums without you, it's probably something else. But that was a stable diffusion or. Sorry, Social Media examiner. My bad.

Yeah, yeah, that was actually a more recent podcast, the one I did with Mike about training your images. I don't know, probably came out last year sometime, I think.

Okay. Actually, I think I found it here. Yep. How to generate headshots with AI.

Yeah, yeah, that's. That's the one.

Right on. Yeah, we'll link that up here. Beautiful. I figured that would be the answer. Anyway, I just wanted to ask it because it was top of mind.

Interesting enough though, from our split testing. Now, if you look at my YouTube channel, you'll notice that I'm actually using my real face instead of an AI generated face a lot more often because the split testing. So when it comes to, like, what images work for thumbnails, it's been a moving target this whole time. Right. And we find that out through split testing Thumbnails is that a year ago, me putting my AI generated face in the thumbnails is what really, really worked. But then we saw that Trend happen on YouTube, and more and more people started doing AI generated versions of their face. Now what works on my channel is actually real images of my face tend to get better.

Optimized, better with the thumbnail test than when we use AI generated images. So we're constantly testing back and forth. And right now we're in this phase where my real face works better than.

My AI face marketers have never been so spread thin. You know, like, with all this stuff happening, technology is going crazy. It's really tough to really focus and get creative and, and create great content or connect with the customers like we used to or even prospects. So, you know, you're trying to launch a bunch of content campaigns, you're generating leads, you're trying to score the leads, you're nurturing them so much more, but you really don't have enough time to even breathe, let alone do the best marketing out there. But HubSpot is trying to fix that for you. They have a new AI, it's built into the system called Breeze. So it actually combines their marketing hub and, and the content hub with all these different cool features that help you automate and, you know, remix your content into a whole bunch of different types, score your leads in a better way, and actually have a nice look at the analytics and everything that's going on in your business. So I would recommend to stop spreading yourself out so thin and start making some big moves with HubSpot. You can try it out for yourself. Go visit HubSpot.com/marketers to learn more. I want to pause for a moment and shout out another podcast on the HubSpot Podcast Network, your audio destination for business professionals. This one's called OG Mentors. It's hosted by Julie Mayer and Jimmy Edgerton. And these two, they're a pretty cool duo. Where one is, you know, Jimmy is a lot more on the loose side. I would say that Julie brings it in and keeps us focused on mentoring and movement. And so these folks, they, they focus on mentoring and exploring ways on how to augment and refresh, kind of like the way that mentoring is applied. And they have these principles that they weave into their episodes and they, they give a bunch of practical tips and they lace it with a whole bunch of humor as well. It's always cool when there's a cool duo happening there. So if you're in the mentoring space or if you have a, you know, leadership space, relationship guidance and things like that, that's what the podcast is made for. So you'll get a lot for it. It's great for entrepreneurs, marketers, sales professionals. They love the show as well. It's a great one. I've checked it out and I recommend it. So go listen to OG Mentors wherever you get your podcasts.

I'm going to do a quick shout.

Out one more time for the thing.

I talked about, the top of the.

Episode, which is the AI version of.

Me and of this podcast and my.

YouTube videos and all these things. It all lives within this AI platform.

I used Delphi AI to create it.

And I want you to use mine. It's totally free. Go to hustleandflowchart.com AI Joe. That's hustle and flowchart.com AI Joe. Test this thing out.

It's crazy. You could type with it, talk with it. You can even use video, and you.

Can look up anything from this podcast.

You can ask a question about your business life, any scenario. It kind of learns about you with the information you give it. It's rad. You will love it. And also, there's a lot of people.

Who are asking, how do I get one of these?

How do I build it out?

The cool thing is my team and I have been building a ton of.

These every single day. I've used the platform Deli AI for two years. I'm actually one of their advisors because I was one of their first users and gave them a ton of feedback. I love it, but we build these things, so if you want to use it for lead generation, for, you know, you have all this content as an.

Expert, author, coach, type person, we can put that into this digital brain and make it interactive. You can use it for customer service, internal training for your team, for your organization.

Lots of uses.

If you would like one of these.

I can pass over an exclusive deal. There's no one else that's. That's offering this out there. I work closely with Deli as one of their partners, agency partners, to build this out. So hit me up.

Type hustleandflowchart.com interest into whatever browser.

Again, that's hustle and flowchart.com interest and hit me up. Let's build this thing together. All right, back to the episode.

So I have a guy named John on my team who makes all my thumbnails now. And so he's also the one. He's like. I guess you could call him like, my head of YouTube optimization. Right. He's constantly going back to old videos and running new split tests on old videos and things like that. So we're constantly, constantly optimizing videos that even came out, like, a year ago. So, yeah, like, even the image you just pulled up is probably a split test that's currently going.

Probably.

Yeah.

And then, I mean, that's a big nugget. You kind of just like, just said, not passing, but, like, going back in time and looking at your history. Sean Campbell said the same thing. He was like, could be Videos four or five years ago that are now starting to pop off or maybe consistently growing. Like, why not go optimize those? Go peek at. I mean, now with all these tools that we have.

Yeah. Then, yeah, we have one video. I don't. I don't remember exactly which video it is, but there was one video that was sort of, like, flatlined. It wasn't getting any traffic anymore. We went back in, change the thumbnail, change the title, and you can literally go into our stats, see the date that we changed it, and it's just like a hockey stick. It just went and just shot up. Right. It was a video that had like 10,000 views over, like six months or something. And now today it's got, you know, 300,000 views after changing the title and thumbnail. You just never know. A lot of times, the algorithm, the YouTube algorithm, nobody really knows how it works, but if it seems like if you go and make some of those changes, YouTube might kind of test the waters with that video again. They might start putting it out to more people and going, all right, there's been some changes to this video. Let's. Let's see what we can do. I mean, YouTube wants the retention on site. YouTube wants people to go back and watch more videos. So it kind of feels like if you go back and change some things, tweak titles and thumbnails, YouTube, I think, kind of goes, all right, let's see if this does a little bit better this time around.

And it's going to give you more love. I want to go back to. Because I could have been hearing you wrong, but the retention on a video, like, does the engagement rate per video, do you think that gives you the most love with the algorithm? Or is it more like maybe they go to multiple videos within your channel, stay on site in general?

Yeah, I mean, I do think it's. I. I think, you know, there's a lot of factors at play, and obviously YouTube's never published how the algorithm works. Exactly. So everybody's just kind of guessing and, like, finding correlations between their tests and things like that. I do think retention on the video matters. I think the point I was making earlier isn't necessarily that you shouldn't focus on retention. It's more that you shouldn't be doing the, like, retention editing style, where it's just these quick cuts and, like, fast motion and, like, try to overstimulate the brain to keep people working. That style of video isn't as effective as it used to be. I still think you need to Keep people around. We still focus on things like the hook in the beginning and opening loops so people stick around to the end and pattern interrupts to, you know, grab people's attention. Who's. When the attention starts to go away, it's just not that like what you would have seen Mr. Beast do a year ago, where it's just like, today I'm going to do this. And then an explosion happens and there's a cut every two seconds and it's just move, move, move, move, move. That style seems to be less effective, but that doesn't mean you shouldn't focus on retention. It just means that that style of editing of just like quick cuts and like overstimulation isn't as effective as it used to be.

That makes sense. Yeah, I figured. I just wanted to make sure it's super clear there.

Yeah.

For any, any of those wondering. Cool. All right, really fast on that other third part of the trifecta. I know there's more, but like, on the email side, I'm curious of some quick and. And then I have a, like a few other questions and I know we have a hard stop. I want to make sure that I can give you some time as well.

I don't need any time. I can literally jump on like one minute before the call.

What's up guys? You don't need a pee or anything? All right, you guys, you do you.

Hey, it worked for me. Damn it.

I like it. This is good. That's the mindset, baby. Email newsletter. What are some, like, you know, some things that might not be so obvious, Obviously grow the list and all that stuff, but like, anything that you've learned now, doing this for what, almost two years? What year and a half or so, like, best practices, maybe new aha's since, since going this route.

So the way I do newsletters now is like dramatically different from the way we did newsletters back when, you know, we had the. The newsletter for Hustle and Flowchart. Never green profits. Right. Because now I'm focused on like more of the traditional definition of a newsletter, right. Where it's filled with the news from AI. Right. Like our old newsletters, we might focus on like one topic and you know, the, the when it comes to marketing, the sort of common wisdom, the common rule of thumb is like one call to action, right. Like one offer, one call to action. Try to drive them to one place.

Right.

And I do think that that's necessary from like a, A sales standpoint. If you're trying to sell something in an Email one like one call to action. Don't confuse people of what you want them to do. The, the style of newsletter I do now, every Single email has 20 links in it. Right. So it's like a completely different style of email and it's just been a completely different way of thinking. It's, it's more about like, how do I provide as much value as possible inside of this email versus how do I get as many clicks from this email as possible? I'm less focused on the click through rate of the email. I want the open rate. That's what I care about. I want people to read the email as opposed to click away from the email to one of the links that I'm sharing. And so that has been like a shift in focus that I've had to adapt to. Right. I've tried to optimize for click through rate but found that when you're sending out a newsletter, certain pieces of news are going to resonate with some people but not other people. Right.

So what kind of open rate are you? Or like what's your target?

Let's just say that I would say I'm targeting a 50% open rate on the emails. I would say more realistically, I get between 35 and 40% open rate.

So the open rates are a lot higher than what we used to get on like marketing emails. Click through rates are a lot lower than what we used to get on emails because there's a lot of options and a lot of people will just read the newsletter and go, that was enough of an overview for me. I don't need to click into any of these links.

I'm not that way a lot of times.

Yeah, I'm okay with that because the monetization is completely different. The way we used to monetize our old newsletters is one link, one call to action. Send them to a site with an offer. If they click the offer, we either make the sale or we make the commission on the affiliate offer. Right. The new way of doing it is get as many eyeballs reading the email as possible and put sponsors in the email. I don't care if they click away from the email. I want them to see the sponsor. Right. So the monetization strategy has been a completely different strategy. Now all I really care about is how do I get people opening and reading the email because that's what gets the sponsor in front of them. So that's been a big sort of shift. And it is really interesting because when you do these like newsletter styles, you do find that the higher up the email you go, the higher the click through rate on the link and then it sort of trickles down and by the bottom, like the last links in the bottom of the email, nobody even clicks on at all.

Right. It's obviously fascinating. Sponsorship opportunity or whatever you want to highlight, you know, where to put it.

Yeah, well, I mean my, my goal is to make sure value first. Sponsors get what they want out of the email second. Right. The value to the user is number one for me. Same goes with YouTube. Same goes with Future Tools. Yes, they're all monetized through sponsorships and stuff, but like making sure the user experience and like them getting the content they want out of it, that needs to be number one priority always. So I always put like the number one most important news thing of the week kind of top and then the sponsor might fall after that and then the rest of the content falls after that. Right. And so that's kind of been the approach at this point. I've hired a team who mostly runs the newsletter for me, but it was definitely like a total mindset shift of how email marketing works from the way we did it as like direct response marketers to now the way I'm doing it is more of a brand marketer.

Yeah, yeah, I like it. And that's. Honestly, I feel like that's just the approach overall, obviously. But there's two camps.

Like you be on the more of.

The media side or be on more of the product creator slash owner side. So there's still a place for media. But of course, you know, you're going to probably direct mainly to your own thing if that's the case. Right, right. So just deciding, but obviously there's whole businesses built off of newsletters now at this point and they're sellable, they're scalable and that's why I really wanted to. Yeah, go for it.

Yeah. That was the other piece with like Future Tools. So Matt Wolf, the YouTube channel, not really sellable. Right. It's too much focused on me. I mean, somebody could potentially buy the YouTube channel, but I would still have to be like an employee of it. Like making the content on the YouTube channel, probably a direction I would never actually go.

Right.

Future Tools, the website and the newsletter is something that if that could be my retirement. Right. Like that's the thing that down the line a company might come in and buy and like what I found about Future Tools is, you know, I've had acquisition offers, I've had a few of them now none of them have been Serious enough that I've like, really entertained them. But what I found is that, like, companies don't necessarily care about the actual website and, like, the amount of traffic it's getting. The. What people seem to want to buy is the fact that I've built this database of all of these tools and like, every single tool that goes onto the website. When somebody submits their tool, I collect their name, I collect their email, I collect the URL of the tool. If they have like an affiliate program, I collect the link to your affiliate program. And all of this data for all of these companies that are building an AI, all of that is in my database. Right. And like, I'm not. I'm not like a data broker. I'm not going to go sell it. I don't want to be the next Cambridge Analytica. Like, I like really a route I'm thinking of going. But the Future Tools website and newsletter is something that is like, sellable in the future. If somebody wanted to buy that company as a whole, I would never just go and sell the database so that people can do cold outreach to those emails. That, like, that's just against my own, like, ethical code. But the business as a whole is something that I can sell down the line.

I think it's brilliant. And the data play. And again, like, this could be applied to so many other verticals and industries. If you're listening to this, obviously you've already outlined how you automate a lot of this on your website and the fact that you are collecting that information. I can just think a bunch of other industries and some that I'm already in as well, that, like, yeah, that data is valuable or like, that's where the money is or like, who else is doing that? Well, maybe no one. Or are they guarding that content? Are there the ways, like, can you publicly. So again, it's like spotting the gaps and then finding, you know, and fill in the gap in your own way. Cool stuff. All right, so moving on. Looking at time here and looking at my notes, I love how, like, we did zero prep for this podcast and it's like I have a whole list of notes now at this point.

But that's just. That's just our vibe, right? We. Whenever you and I get on a call, it's like, there's no way you and I are getting off that call within 30 minutes, right? It's just not possible.

That's why I'm like, I gotta cancel the team call, Jacob. Now you know why I actually called the team. So. All right, there's a few things. Well, shoot. Where do I go? Where do I go now? I want to. I want to go here first. So what excites you for AI? Maybe with what's coming, what's here, what's exciting, and keep it kind of don't go like, I know you can go super deep, but like. Yeah, so I have some follow ups.

Yeah. The thing. So I'll tell you what I'm playing with the most when it comes to AI. I love playing with the AI music generators. Suno and Udo being two of them, like, so much fun.

I cut you off. Suno is one. I've used that. What's the other one?

Udo ud.IO I haven't used it.

Okay.

Yeah, Same concept as soon. Oh, I feel like. So between the two, Suno I feel like makes more enjoyable music. Right? Like, it's more music that I'd kind of bob my head to. And I actually kind of enjoy what it creates. But you can kind of still tell it's AI. Right. You listen to it and the way they sort of blend words together and they still haven't figured out how to say AI. A lot of times it'll just go I right. Like.

Like, that's good. So Suno, like, you'll listen to the songs and be like, okay, I could tell that's AI. But it's actually pretty good. I kind of like this song.

Right?

Yu D O is a lot better at fooling you into thinking it's a real song. Like, I will listen to that music and be like, this sounds like a real band. I couldn't tell if you played this for me that it was AI cool. However, I feel like the music's slightly less enjoyable. Like, I'm not finding myself like, oh man, this song freaking rocks. Right? Like, I don't. I don't feel that way as much with what Yu Dio generates. They're both good. They're just good for separate reasons. The other stuff that I'm really excited about is AI video generation. Right. I'm sure you've seen like Luma's Dream Machine where you can enter a text prompt and it'll generate a video off of it. It doesn't really work very good for text to video, but it works really well. If you generate an image with Mid Journey or Stable Diffusion or Dall E3, pull that image in and use that as like the starting frame of the video. Amazing at that.

That's a good tip right there. Okay.

Yeah. And then Runway also sort of teased their Runway ML sort of teased their Gen 3 model, which is another sort of like SORA level text to video AI generator, which looks really, really good. So when I'm just like sitting around having fun playing with AI, I'm playing with the music generators and I'm playing with the video generators and a lot of times I'm doing it like in tandem. I'm generating a song and then going to like Suno or generating a song in Suno and then going to like Luma and generating like a whole bunch of video clips to go along with the song and then making a music video using the song and the clips that I generated with like, to me, that is so much fun. Like, I love that stuff. Like that to me is really, really cool. Claude keeps getting better and better and better. I love using Claude. I know.

Released what, Sonnet 3.5. Yeah.

And it's. And it's outperforming GPT4O already. Yeah, it's out like in all the benchmark tests, it's outperforming. It can do a single prompt. Video game, right? I've saw, I saw somebody prompt it to like make me a, like a basic game, like make me a Snake game. It coded the whole thing and it worked right out of the box. I did a video where I was like, make me a playable Tic Tac toe game where I'm playing against the computer. Made it with one prompt. The game was workable, playable right out of the box. Now no other AI is really doing that right now. So Claude is really, really kicking ass for like writing code for you right now. That one's been really, really cool. I, I still use Chat GPT on my phone a lot. Like I, yeah, I, for whatever reason, I like the Chat GPT app quite a bit. So I use ChatGPT on my phone, but when I'm on my desktop I use Claude. But yeah, those are the things that I'm playing with that sort of excite me right now in this, this current moment.

Is there anything coming up on the horizon that, because this is like the video thing, I felt like was on the horizon for a while.

Yeah.

Because of. I'm blanking on the name. The open eye. Sora. Yes. Yeah, I was going to say soon over like, nope. Sora. Anything coming? That's exciting.

Well, we're still waiting on Sora, right. Luma only generates five second videos that Runway gen 3 can generate up to 10 second videos. Supposedly Sora can generate one minute videos. Right. So like that additional length from Sora Once we get that, that'll be really, really exciting. You know, the stuff that I think I'm the most excited about that's like in the pipeline is more of the like agentic stuff where you're building like these agents that will actually go and do tasks on your behalf. Yeah, right. So like I work with a company called Mind Studio. They help you build like these little AI agents as well.

That type of stuff is, is really exciting. Like, you know, go and do this research for me. Come back with the research. Once I've, you know, based on your research, go find the product on Amazon, price, shop it for me, find the cheapest place I can buy the product based on your research and then, you know, send me a confirmation and I'll give you a yes or a no. If I say yes, then buy it on my behalf.

Right.

Like some of this agentic stuff is sort of popping up and we're going to start to see that more and more and more in the coming months of like these, these language models that can use tools and go and actually take actions on your behalf. I think that's what we're going to start to see bubble up more and more and more over the next several months.

And that's where like, you know, I'm doing a lot of stuff with a lot of clients, enterprise level, you know, with Mike Koenigs and Brad Constanzo and you know, just across the board I'm seeing a lot more people starting to think that way.

Yeah.

With the agents, I think people now are understanding at least the basic fundamentals of chat GPT, Claude. They might not know really how to make the most of it, but at least they're already starting to think, oh, what if I string these things together and that's exactly what you're describing in Mind Mind Studio. Got to connect me with the founder, by the way. Yeah, yeah, yeah, he's my neighbor. Our neighbor.

Yeah, yeah.

He lives out in. Yeah, yeah, yeah, yeah. Hey, don't call him out like that.

Oh, sorry, yeah, my bad.

It's only horses and a bunch of lifted trucks out there.

Right.

We're in east county all so. But no, the Mind Studio seems like the best one I've found because I've looked around for solutions and yeah, for exactly what you said, like it's, it's a sweet business like use case if you're looking for a lot of these agents and yeah, it just looks fun. I personally haven't used it yet.

I need to though, I'll share one more thing that I think is that we're going to see a lot more of as well. That I think is exciting is more of like the wearable tech, you know. Yeah there was the things like the rabbit R1 and the humane pen and you know they just got fricking destroyed and all the reviews online and I don't think that's it. Like I never thought that was it, right? Like why have a separate app, a separate little handheld device when my phone does all the same crap already, right? Like it didn't make much sense to people.

Intelligence.

Well yeah, and Apple Intelligence is going to basically killed rabbit and humane in friggin one update, right? But what I think is more useful is things like, like I've got the Meta Ray Ban AI glasses, right? There's been some demos of like EarPods, right? And the AirPods they look like the Apple AirPods but they have little cameras on the tip that are like 360 cameras that can see all around you, right? And like based on what you're seeing can sort of feed information into your ears. And so what I think is, is going to happen is we're going to see more and more of these wearables with AI built in. And right now the AI that's built into the glasses isn't very useful, right? You got to talk to your sunglasses and say hey Meadow, what am I looking at? And it'll be like it looks like you're on the beach and it's like cool, what's that good for? Unless I'm like, unless I'm like hard of seeing or blind or whatever, right? And I, and I need the glasses to sort of give me a description of what I'm seeing. It's not really that useful to me. But what I think is going to happen is they're going to start putting like little heads up displays in it. Or.

You know I was actually, it's funny that we talked about Mind Studio. I was talking to Dmitry, the CEO of this of mine studio and one of the things that we were, we were really sort of nerding out about is the killer app would be sunglasses that when you meet somebody, remembers that person and then the next time you bump into him the sunglasses whisper in your ear. By the way, this is Joe.

You know, here's a couple of nice facts about him.

This is Joe, he's got two daughters, he's married, he's got you know like, and it gives you some like information. He lives at.

He lives at 1-234-S-DRIVE, you know, whatever, whatever Information it remembers about this person. You walk up and it's either like giving you a heads up display on your glasses like this is Joe. Or it's like, you know, telling you something in your ear about Joe. Right. Like that's cool.

I can see that would be like.

A killer use case. Like be my. Like how many people are saying they, they're horrible at remembering names. Well, put on a pair of glasses and now you remember everybody's names. Like I think that's coming soon.

Yeah. And follow up, like you could probably initiate follow up like that instantaneous. Like with all these next steps that you might be doing when you're meeting someone. Like, yeah, it's. Yeah, I think it's, it's just like how AI already is here on the screens. Why not bring it into the world around us and make it visual, make it here, make it all. All the above. So I. Yeah, 100. Cool. It's got to start somewhere with those meta Ray Bans. I mean it's like it's going to start basic. It's like Google Glass or whatever. The glasses.

Yeah. Yeah. They were ahead of their time, really. Like now they probably would have. I mean they, they're ugly as hell, but now they would probably be a little, slightly more accepted than they were back then for sure.

For sure. All right, go. What scares you?

Quite honestly, I'm not that scared of most AI stuff at this point. Right? The, the stuff. Okay. So the stuff that really scares me around AI is more like bad actors using AI to do bad actory things, right? Like using AI. Like we talked about Claude and how Claude is really good at coding with a single prompt now. Well, if Claude is really good at coding with a single prompt now, how easy is it for anybody to become a hacker, right? Like if I can go and have one of these AI and you know, Cloud and OpenAI, they're putting guardrails on it. So people can't do that. But there's also open source models that are quickly becoming just as good as the closed models. Right. And if you're using an open source model, you can yank whatever guardrails you want off of these models and do whatever you want with them. Right. So what happens when these open source models get good enough at coding that they're as good as Claude and I can say, write me a script that's a trot and horse that I can install on Joe's computer and steal all of his data. Right. Like.

You'Re probably capable. Don't do it.

Or like you Know, the. As the video and image generation gets better and better and better, how is, like visual evidence ever going to be used in court anymore? How are we going to show a video of, you know, somebody getting attacked and prove that that was. That really happened versus generated with AI? Right. The AI voice cloning is getting really good. We've seen scams already where people have used other people's voices, called the person's parents, scammed them out of money, trying to be somebody else because they were able to clone their voice with AI. So the stuff that like bad actors can use AI for, to me is the most scary thing about AI. I don't buy into the scenarios of like, them rising up and destroying the world, at least not with large language models, not with the AI that we're using today. They're not even capable of that. There's, there's almost like.

There'S really no path to that with our current, like, tech that's out there.

It's pretty dumb tech still in the.

Grand in that sense. Yes. Yeah, but the, the sort of scam potential, the deep fake potential, the, the hacker potential. Yeah. You know, there's already been reports of Russia and China and stuff like that, meddling in elections, using AI and you know, who knows if. If it's propaganda or reality. Hard to disseminate that stuff now. Hard to determine the truth versus not truth. And AI is not making it any easier.

It's getting more complicated, more. I, I keep thinking of, and this isn't the scary part, but like, we're in a time of.

Almost like a.

New renaissance in a way. And you know, it's like there's different arts, there's different media, there's different things that we all get to explore and kind of reinvent ourselves, reinvent society, you know.

Yeah.

There's obviously a lot of scary stuff that can come from that, but it's like it's new, I guess. So it's like this whole new reality, I feel like, is shaping up rapidly in front of us. And it's, it's adapting to that. It's like a different way of thinking that a lot of us, all of us need to.

Well, yeah, and I think, I think one of the big struggles we're going to run into is like governments are going to try to regulate this, but they're trying to regulate based on, like, the way things used to be with new technologies. And that just isn't going to work. Right. Like, I don't think copyright law is going to exist in the same form that it exists right now, 10 years from now. I just, I don't think it's going to be possible. You know, like.

There'S all this, these sort of issues about like where the training data came from on like the large language models, the video models, the image models. Like it's trained on other artists work, it's trained on the writings of other people. Right.

All the books out there basically.

Yeah. And that really sort of muddies the waters a lot. But if you look at humans in general, they operate in the same way. If I become an artist, I learned from all the artists that came before me. I'm taking influence from, you know, the artists that I tried to model when I was learning art. How is that much different? Right. You know, I actually have this belief that all of the people that are fighting against AI over copyright reasons are actually pushing AI to become more and more mainstream. I think the people that are fighting it are actually having a reverse effect from what they're going for. And so you look at a tool like Suno that can generate AI music. Right. Well, in, in the old days you would have to go and license somebody's song, put it over your YouTube video, and if the song didn't happen to be licensed, they would slap you down. The video would either get removed off YouTube or all of the monetation, all the monetization on that video would go to the creator of that song. I could have a 30 minute video and a 5 second clip of a song and if that, the copyright holder saw that my, that five seconds was in my video, they can take 100% of the revenue from that video even though it was only five seconds of a 30 minute video. Right. So what does that make me do? Well, I'm not going to go license other people's songs. Tools like Suno are out now. I can just make a song, not have to worry about licensing it and put that in my video without worrying about any repercussions. Right, so by you going and slapping down YouTubers for using their song in their videos, well now people are going to alternate sources to do that. Remember when we used to do like blog posts and you know, we might find an image on Google or something, throw it into our video and then you get an email from the Associated Press saying hey, you owe us $800 because you used that image on our blog post.

I do, yeah, I remember that.

What do you think that's doing to people now? Okay, fine. I am never going to use an Associated Press image ever again. I'm going to go generate an image with midjourney instead.

Yep.

Right.

That's right.

These companies that are enforcing copyright law are pushing people to the AI alternatives, which is not helping their cause. I just think copyright needs to be rethought completely. If the creators of this, this IP want to continue to like, actually be relevant in the creator economy and think.

About the people who are pushing back and all that, it's like, well, you got to, you got to know technology is not stopping. It's going to keep moving quicker and quicker. It always has. And maybe, maybe the better stance is to be that person that helps shape what regulation, new regulation works like. But understanding that technology is going to keep moving forward, someone's gotta take the realm and like, if you're specialized in, yeah, music licensing, it's like, great, you know, own that and then be the person to help make that a thing. That's, you know, it's helpful for everybody as a whole. It's probably going to look different though. It's not.

Yeah, I mean, Grimes had the right idea, right. You know, Elon's one of Elon's baby mamas, right? Like, she had the right idea of going on, on X and saying, hey, you can use any of my music you want. You can use my voice, make new songs with my voice, do whatever you want with it. Just give me a percentage of whatever you generate from it. If I was an artist, I'd be saying, screw copyright. I'm going that direction. Here's, here's the files to use my voice. Go make songs with me in it and give me a cut. Look, I just created an army of me's now. Like, that's a better model. I'm sorry.

Kind of brain.

Yeah, I'm getting all the marketing love now and the first mover to do that, or at least handful, like, are going to win.

Yeah, yeah, exactly.

These are going to be known. All right, so on a, on that note, and I'm looking at the clock still.

How are you staying with this path? Like, I'm thinking more mindset now. And because you're, you're going hard. How many. I mean, at least, what, two episodes or episodes. Two videos per week, right?

Yeah, I try for three on really busy weeks. I end up getting two. But I would say the average week I'm putting out three videos.

So you're obviously like, you got the website pretty damn dialed in. There's obviously still a filtering process. We don't need to go through all that. But you automated a lot of the Work website side. You have someone else writing your newsletter for the most part or completely. Videos though, and staying up with the news. And then now I'm thinking of the sponsorship obligations you have across the board. Partnerships. Obviously you're having the YouTube or, sorry, podcast side of things. So you're creating content there as well. Yeah. How do you sleep?

Do you.

I mean systems and teams. Right, Systems and teams when it comes to YouTube. Right. I've got, I've got John who helps me with all the YouTube optimization, the thumbnails. I've got Mark, who's my editor. I still edit a lot of the videos myself because I just enjoy editing.

I figured. Yeah.

But I still send, you know, I'd say half the videos Mark edits, half the videos I edit. So I've got an editor who's helping me there. I've got a team who helps sell the sponsorships for me. I've basically got like an assistant who makes sure that like, all right, don't forget this Wednesday you've got to put this sponsorship into the video and like, helps keep me on top of that stuff. So it's really just a lot of team, a lot of systems and that's how I'm able to like crank out as much as I crank out. Like I've probably got, not directly working for me, most of them are contractors, but I'm probably working with a good like 15 people on a daily basis now. And I've even gotten to the point now where like I've got people managing the other people. So like, so I've had to figure out how to build these systems to keep everything going. But like getting into the sponsorship stuff, that's probably been the toughest part of like this sort of transition into YouTube for me is because.

So when I started doing a lot of sponsorships, I think last time we talked, I talked about how I signed like a lot of long term deals with some partners and I've got like one, two year deals with a lot of these partners. Well, back when I signed those deals, my channel was smaller. I was getting less views per video. Well, as the channel has grown, a lot of the audience has become less and less receptive to sponsors. Right. So like some of the deals that I made included like dedicated videos, for instance, where I would make one video and the whole video would be about this one company. It wouldn't be like, here's all the news, by the way, this video is sponsored by X and then move on with the news. People don't seem to mind that. But if I make a video and the whole video is about one single company, people hate it. Right? Like, the comments are all like, oh, you're ruining your reputation. Oh, you're shilling for this company. Oh, you're. And I'm like, damn it, man. Like, I signed this contract with them like, nine months ago when people didn't really seem to be paying as much attention to my channel as they are now. And now it's like, all right, I've got to fulfill on what I promised in the contract. But the channel has gotten to this level where people seem to have a problem with that. And it's like this constant balance. And now I'm going back to a lot of the sponsors that I have, like, dedicated videos with and trying to renegotiate old contracts and say, hey, instead of a dedicated, can we do three integrations instead? And they're like, we'd really prefer. And so a lot of that is like, the struggle for me. And I'm not going to name any particular sponsors. I actually do really enjoy working with all the sponsors I work with. It's just this, like, it really sort of weighs on me that I want to make content that comes across as unbiased. That's just sharing the news that shows my excitement. And then when I got to work these ads into them and sort of like shill for somebody else for whether it be like a minute or like an entire video, I just feel so inauthentic doing that. But it's like part of what makes the money doing this. So it's like, I don't know, like trying to find that balance has probably been the biggest struggle I've had so far.

Yeah. Yeah. And I feel like you're always. You are always authentic, as far as I can tell, you know, on the video side, I mean, I don't know all the nuances of the tools and whatnot, but it's. It's tough. And it's not the sponsor's fault. It's not. It's no one's fault. It's just kind of, I guess what it is. Right.

The nature of the game.

Yeah, yeah. That's really what it comes down to. And obviously it's like, I don't know, it's a. It's a tough game and there's no perfect solution there. But like you said, it's like you're. You're juggling that multi year sometimes. Like it's. It's a good and bad thing, I guess. Right?

Yeah.

As a creator. Yeah.

It's interesting. I feel like the multi year contracts was like a really, really, really good idea when my channel was smaller. But the bigger the channel gets, the more you're like, okay, well, you got in on this deal back when I was getting way less views. But I mean, that's part of why you sign the long term deals, right? Is you get in with the sponsor, wants to get in with you, seeing a rising star or whatever, and like, hey, cool, I'm gonna. If I capitalize on this now, then I get them when they're a lot bigger too. So, like, I under. I understand that part of it. I just, I don't think I anticipated the trolls and the backlash and the negative negativity that would come from those, that kind of content later on down the road. Like, I didn't sort of anticipate that happening.

Does that ever get to you, like the trolls or, you know, comments like that? Because, you know, Rogan, Joe Rogan would say, like, don't read your Twitter comments.

Yeah, I don't really get much like trolling on Twitter or in like the, in the, in the communities where I engage. It's mostly positivity. Most of the negativity is in the YouTube comments, which I try to avoid, but I still catch them from every once in a while, right. Like, I still have to go into my YouTube account and I still see comments in there, Right. If you go to your dashboard, YouTube doesn't even let you hide it. The comments are right there in your dashboard, like the most, the three most recent comments, right. So every once in a while I pop into my dashboard and I see something just like horribly negative. And yeah, it affects you. I don't. Like, I, I listened to a Casey Neistat interview recently and Casey's like, dude, I've been doing this for 20 years now and it still affects me when I see like trolls and negativity. Like, I don't know if you ever get over that, but you do learn to have like a thicker skin around it, right. Like, I, it, it affects me for a half hour and then it rolls off and I forget about it. Right. Like, the amount of time it takes me to like, forget about it is a lot quicker than it used to be, I guess is the best way.

Of putting, Shortening the gap. Yeah, it's good, man. That's great. Well, I mean, we can, we can go further on the, I think.

Yeah.

On the mindset side, like, is there anything else that you think, like, you could. I'm just thinking if you Started all over. Like you tell yourself like how to prepare for what you're in the middle of now. I don't have a perfect question, but some whatever lands there.

Yeah, well, so one of the things that I've realized is most of the stress and most of the pressure that I've got from YouTube is all put on by me. Right. Like I put the pressure on myself to try to produce three videos a week and I try to put out three videos a week and the weeks that I only put out two, I sort of beat myself up for it. But I don't have a boss somewhere sitting there going, why didn't you get that third video out this week? That was just my own like constraint that I built for myself. Right. Like so a lot of the like pressure and stress is like, is self inflicted did. Right. I agreed to all of these sponsorships, I agreed to all of this stuff. I think, you know, if I was to give myself advice I would say, you know, don't worry too much about the like the frequency of the content. Right. Like worry about putting out really good content in the beginning. I think.

You know, I don't know though, that's hard because that's sort of the advice I would give myself today, not the advice I would have given myself when I first started the YouTube channel.

Because you needed frequency to get the ball.

Yeah, the frequency is sort of what kickstarted the channel.

Channel.

But then I, I almost feel like I got to this point where I felt like I needed to keep that same momentum even once the channel was already big. But the bigger the channel got, the more I can get away with producing less content but going higher quality. It's almost like the bigger the channel gets, the more you need to focus on quality versus quantity. It's like an inverse relationship kind of thing. You know. I don't know if I would do a lot of long term sponsors. I don't think I would ever do dedicated videos. I think I, I have already said no more dedicated videos. We don't sell them anymore. They're not an option. I still have a few that have already been sold that I need to fulfill on, but I don't sell them anymore. And I think I would have given myself that advice. Don't do dedicated videos. They're harmful to the channel integrations. Cool. Dedicated. No, got it.

Makes sense. Yeah.

But yeah, I would say though that's like, I don't know, it's, it's hard to say because I, I, it's hard to put myself back in the mindset that I was, you know, to two years ago when I was first kicking this all off and started going down this path. But I do think the free, the high frequency, maybe not as high of quality is what helped me in the beginning. Then as I got bigger, I sort of shifted more into quality versus quantity. And yeah. When it comes to sponsorships, pick and choose the sponsors. Right. And don't do any dedicated.

Yeah. Yeah. Know what you're getting into and thinking. Yeah. It's almost like not always solving just the now thing, but thinking ahead just enough.

Yeah.

So, yeah. All right, amigo. We are two till the time you get a bounce. So I took you to as far as we can take you.

Took it to the limit.

Thank you so much. Shout outs, obviously. Future tools IO find Matt Wolf on. On everywhere else, basically, but YouTube. What Mr. E. Flo on X, Twitter X.

And Instagram. Yep.

Instagram. Yeah. Join us.

Don't really do a lot of Instagram. I'm trying to do it a little bit more, but.

Time. All right, man. Appreciate you so much. Love you, bro.

Cool, man.

Likewise.

Appreciate you. Love you. And thanks for having me again.