Matt Wolfe - Top 10 AI Tools All Entrpreneurs Need To Be Using

We are back with episode number two with my previous co host, Mr. Matt Wolf, now of FutureTools IO, and he's writing his cool blog post and all that stuff@mattwolf.com this is the second episode of Two, and so we're gonna do some recurring things maybe every six weeks or so, but we have kind of a two parter. So last week we put out the episode of basically what the heck's been going on since us splitting up as business partners, as co hosts of this show that, that we built for six years up to this point. And now I'm taking it over as solo host and I'm bringing it back on for the second episode to basically talk like go deep dive into AI, artificial intelligence, machine learning, what it is. He kind of corrected of some terminology that I was using, but also that everyone uses. And so you really understand where this stuff is going in terms of tech. You're seeing a lot of stuff right now in the AI side of things. There's a bunch of tools being released. There's all these different things that you can now do to help your writing, to create music, to create videos, to create art, and to do all sorts of different things, edit your podcast. We're going to dive into all these different tools that Matt is loving, but also kind of what this thing is like, why is it so important how it works and at a high level. And then also where is this thing headed with a whole bunch of practical uses throughout. So enjoy this conversation with Matt and my brother from another mother. Here we go. Hey, I'm Joe Fear and welcome to the Hustle and Flowchart podcast. This is where we talk all about building businesses so they give you the freedom and fuel for your life. I'm not here to help you build a billion dollar business, but I am here to help you create a business with systems that work for you so you can make more money than you need just by working part time. You know, I was a chronic hustle mode kind of guy and I want to share my experience, experiences and mentors I've met along the way to help you reframe things to be the most effective as an entrepreneur. I wish I had this guidance and insight when I was younger, so that's what I'm doing here for you. Please share and enjoy. Hey, we're back in episode number two, numero dos. Hey, Matt, how you doing?

Hey, long time no see.

Yeah, it's been a while, huh? So if it's been a while for you guys being like, hey, Where's Joe and Matt been? Matt and Joe, however you think of it in your mind, go listen to the episode one because that explains, kind of brings you up to speed. It's not too long. We're keeping these short and sweet for y' all as best we can, and.

We'Re not making any as best as we know how.

That's all we know. So this style of episode is going to. Or this episode number two is going to be diving into AI and the crazy tech that AI is and how many different tools and things that are coming out to market. Seems like multiple a day. And these things are creating just so many ways to just prompt AI and then it spits out an essay, a recipe, or whatever you want. Helps you with your email, your grammar, your music creation. I tried some of that, Matt.

Yeah, it's cool.

It's pretty fun. What are some other ones? It helps you edit podcasts. I'm using a couple tools that you. Time Bolt's one of them, and Wisecut's the other one. I'm thinking Time Bolt because it's more visual and you can actually act like an editor, whereas the other one's almost like hands off editing. That's.

And also, also, yeah, shout out to Time Bolt. They listen to this podcast. Oh, sorry, I've been emailing with the founder.

Oh, Time. Okay. Well, that is my genuine review of Time Bolt so far. I've not paid you money again, sir, but I will. But we should chat actually, on this podcast, too.

I put Time Bolt on future tools and they actually reached out to me and said, hey, I've been listening to your podcast. I love you guys.

Oh, my God. I'm using Time Bolt. That's what I'm doing after this. So this episode was likely edited with Time Bolt. Go check it out. I don't know if that's AI or whatever, dot, AI or you find them. But the point is there's cool tools to save you a ton of time and. And automate things, create things out of thin air, help you with gift ideas. I think, Matt, you have some things that you're going to share with us on that, you know, like, it just kind of allows you to, I don't know, tap into your creativeness in a way that there's obviously a lot of controversy with AI, especially when it comes to art and artists, but let's be honest, all tech will always do that. It always upgrades us, and we don't even know where we're headed. And it's going to create new opportunities that we never even thought of before. So I want to. So we're going to deep dive into that.

Matt, let's talk about all of it. I love the topics. I love talking about the ethics of it. I like pointing out tools that nobody's talked about. I love sort of making predictions of where I think it's going to go. All of that is fun to me. And I might be completely wrong and off on all of it, but it's still fun to talk about.

Well, I want you to throw it out there so we can, we can fact check you in years to come. We're not going to remember to do that, let's be honest. So now what I have written down here is kind of what you said. I like the applications of this, of these tools. Like, I really want to. I want to talk about the landscape. So maybe we start as like an overview of your thoughts of what it is. Not too deep on the landscape because let's be honest, everyone kind of knows what AI is.

But.

But explain it for like in terms of people wanting to apply it to their lives, be it in a business or maybe just their lives. Like I was saying, looking up a recipe for an instant pot recipe about chicken enchiladas. I looked it up the other day and I'm like, it was better than any recipe. I googled and found a blog. Cause I got to deal with all the stupid ads and you know, the bio about the chef. And like, I didn't come for the chef, I came for the recipe. I know you need to make Money, but sorry, ChatGPT, just one out on this race.

Yeah, yeah, yeah.

So let's start with the landscape and then we'll dive into what some of these maybe the top like tools, but with the idea of application and using these.

Yeah. So I think. So first of all, let's start with like a definition because technically none of this is actually AI. If we're being real, we're not actually using AI. What we're using is machine learning. So everything, all of the tools that we're talking about, let's say, you know, the generative ART tools, the ChatGPT tools, the generative audio tools, all of that, it's all technically machine learning tools. It's all been trained on a data set. And that data set is what it's processing everything off of. Right. So ChatGPT, for instance, is like the really, really popular one. It's trained on a data set that only goes through 2021.

I was going to ask that. So that's what I thought. I thought it was actually a little. But okay, 2021. So you. So you're not going to get up to date stats on anything from 2022.

And oh, and so it doesn't go and search the Internet for you. It doesn't go and do any of that. It's got a data set that has been trained on and everything that it's submitting back to you is what it understands from the data set that it's been trained on. It's technically machine learning. If we're talking about true AI. True AI is going to be self learning, Right? True AI is going to be chatgpt. And if it doesn't know an answer to something, it's going to go and figure out the answer and, and then teach itself the answer and then continue to further and further train itself.

It would look up a YouTube video, it would digest all the information, digest that, program that, and give you an answer that's up to date to the moment or whatever. It might be a data set.

Yeah, it wouldn't need to be trained on data sets. It would go and be self learning, essentially. Right. But AI is such a better buzzword if you went around and said, hey, this is a cool tool that does your copywriting using machine learning. It just doesn't sound as cool. Right? So everybody uses AI and machine learning interchangeably, but we're really talking about machine learning here. AI is much, much scarier.

But, well, we'll get there later, I guess. Thank you for the clarification. So machine learning. No, it's great because I do want to know this and so any other. Yeah, keep, keep going. But I love how you start here.

But yeah, so like the general landscape of it right now, and it's interesting because I saw a post on Facebook today saying, man, I don't know about all this AI stuff. I feel like it's just making people dumber and lazier, right? Like, hey, I don't want to do my own art, so I'm going to go throw it in here and it's going to give me my art. I don't want to do my own copywriting. I'll just throw a prompt in here and it's going to give me some copy. Like AI is making people dumber and lazier. And I went and responded to it going, huh, I've had the exact opposite experience. AI has been making me smarter. You know, these tools have been making me smarter because I went and used ChatGPT to start learning how to code in Python. I wouldn't use ChatGPT to go and understand terms that, that people are throwing around in this space that I didn't understand yet. Like you can go to ChatGPT and go, what is machine learning? Explain it like I'm five, right? And it will explain it to you in a way that's easy to understand. So in my mind I am using these AI tools to make myself smarter and to use them as like a supplement, a compliment to my brain and to augment and actually make the learning better and easier. So, you know, but so the general landscape right now we're seeing a huge, huge spike in generative art, right? There's a bunch of tools in the generative art. It kind of started with dall E from OpenAI and then stable Diffusion, which is more of an open source one which you can actually install on your computer and train your own models and do really cool stuff with. And then there's Midjourney, which in my opinion is actually the best generative art. It actually makes the coolest looking art, right? So you've got these generative art tools and then you've got these various tools that are leveraging the GPT3 technology also from OpenAI. ChatGPT is leveraging this GPT technology and sort of turning it into more of a chat communication. But you can Also go use GPT3 and just plug in like text and say, hey, summarize this text for me. You don't actually have to use ChatGPT to get that. Those are the two kind of things that we're seeing like really bubble up are the generative text and the generative art. But that's just like the tip of the iceberg. Which is the reason I created the Future Tools website was everybody's talking about this, but nobody's talking about all this other really cool shit that's out there, right? There's generative music, which you hinted at. There's one called Sound Draw, there's one called Muse that's being created by OpenAI as well. And tools like that, you can literally go and create royalty free music that is unique to your video. It's never been used before in any other video. You can say, I want it to be dramatic and I want it to be slow paced through this portion. And then when it hits the 45 second mark, I want the beat to drop and I want it to get faster. And then when it hits the, you know, minute and a half second, I want it to slow down again. And then I want the video, the song to end right at three and a half minutes because that's the length of my video. Right. And you. And it will go and generate a song that meets the criteria that you went and put into it. Right. Nobody's talking about that stuff yet. There's really, really cool text to speech generators out there. A lot of them are using Microsoft Zero, which is sort of a backend API that you could use. And a lot of them are leveraging that API, but it's these text to speech generators where if I was to type something into it and play it back to you, you'd probably have a real hard time telling me that that wasn't a human saying it back to you.

So I'm literally@FutureTools IO right now looking up and you have this like this filter feature that's so simple. And I just typed our selection, selected the speech to text. Actually it's Text to speech. Huh? I went backwards. Yeah.

Well, there's both. Speech to text would be like the script and tools like that. And then text to speech would be like the Microsoft Azure and there's a whole bunch of tools that will actually convert your text into speech for you now.

Yeah. Murph, AI like you said. What Play ht?

I think Uber Duck is an open source one.

Yeah, Uber Duck. There you go. Text to song.

Yeah.

Use AI to text and turn it into a song. Like this is wild, man. You can go down the rabbit hole, which I know you are. You probably get no sleep anymore.

So. Yeah. So that's where exactly where I wanted to head to is. Okay. You laid out the foundation of AI is really machine learning. But AI there is evolutions to what we're seeing right now. It just so happens to be like in December 2022, ChatGPT has taken over the airwaves. It seems like. I know their site constantly gets bogged down and it's free. I think just in this research phase, I'm sure they'll token. I mean they're. I heard something like they're. It's millions of dollars per day that server costs to run. All 3 million a day, I believe. Yeah, she's. And I'm sure maybe it's going up like it's probably still low. So you know, they're going to put a. I don't know. There's deep pockets obviously, but still like they're getting so much data. That's what's so interesting is they're mining the data. Everyone's going there. You mentioned to me like this is Google killer, a Google Killer talk about that and how, you know, maybe starting with what we know of, like more in the public sense, it's been chat GPT. A lot of people are using it for things, but you know, like, there's all these different other tools. So why Google Killer? And then like also what is coming next in terms of like, what's the mainstream AI thing gonna be? Like the evolution?

Yeah, well, it's funny because you brought up the monetization and how they're spending $3 million a day for the servers. Well, Sam Altman, he's the, he's the CEO of OpenAI, he was recently interviewed and in the interview they said, hey, I know this costs a lot of money. How are you guys gonna monetize this thing? And his response was, we have no idea yet. But we're gonna continue to train this ChatGPT and add more stuff and let more users use it. We're gonna improve it and improve it, improve it. And, and our goal is to eventually ask ChatGPT what is the best way to monetize ChatGPT and let the AI tell them how to monetize their own platform. That is a literal thing that Sam Altman, the CEO of OpenAI said about ChatGPT. So I don't think you have to worry about it like becoming a paid thing too close in the near future. I think they want the answer, right?

They want to have the most amount of input from us, the humans.

But I also do know that GPT4 does exist already, right? So GPT4 already exists. And right now, in order to be able to use GPT4, you need to have had an investment from a certain company and then OpenAI gets 10% of the revenue from the company. So there is some stuff going on behind the scenes for people getting access to GPT4 in exchange for rev shares.

What's the difference?

GPT4, which I believe it's just a more advanced, more trained up model. So it's probably trained through later dates and has a larger data set that it's working on off of, if I had to guess. But I don't know specifically, I'm, I'm not in that inner circle yet.

Do you know, like, do they update these databases or is it always like an upgrade to the whole thing? Like instead of 2021 or whatever it was for chat GPT, like can they say, hey, we just plugged in 2022 in there?

And yeah, I mean, I'm not sure. I believe they would have to go in and update it. I would bet they're going to go in and every periodically update it with new data, but I don't know the frequency of how that's going to happen yet. It's too fresh, too new. So who knows? I mean, we're already seeing updates to the user interface, right? Like a week ago you couldn't even go in and like see previous chats. Now it's actually saving all of your chats for you. And a week ago I would, I actually was trying to get it to create a YouTube scraper for me where it scrapes like the, a YouTube URL, a YouTube title and like what keywords. A YouTube video was working and throwing it on a Google sheet and I asked it to create the code for me and it was like, this is against our terms and it wouldn't let me do it. It was basically saying we're not sure that you're doing something legit. And I mean it's totally legit. I can view the source of the page and get all those details right. But now they've made it so it is less likely to give you that response and just do what you ask it. Right. So they are making updates to at least the user interface aspect of it, the data set. I don't totally know how often or frequently that's getting updates, but right now we do know it's only trained through 2021.

Okay. No, this is, this is awesome. Cause I know with ChatGPT, like there's so many applications, if you plug in your prompt differently into that thing, you're going to get vastly different responses. Like a lot of us have used Jarvis for, for copywriting and for doing all Jasper. So yeah, we've used Jasper for, for, you know, creating all this copy and all this stuff in marketing that's been around for what, a couple years or so, I think.

Yeah, it was conversion AI, then it was Jarvis, now it's Jasper.

That's it, man. Yeah, so the evolution of names, but it comes with like the evolution of this whole space. And like with chat GPT, I feel like you could do a lot getting smart with the prompts, but there's no like recipes and stuff like that. But I know you're going to prove me wrong because there's probably all these other things that are tapping. The point is like they're all tapping into all that, like that bigger open AI database. And like you said the other one. I forget what the other one's name is, but like building all spinning out all these other tools that are.

Yeah, well, so if you go over to future tools and there's actually a section called Inspiration, if you click on Inspiration, you can actually see there's tools in there like awesome ChatGPT prompts and then another one called Show GPT which is also sort of recipes for prompts that other people have gotten good responses from. So like I've even curated some of the, some of the actual recipes and prompts that other people have used over there as well. So there are actually like recipe books essentially that people are putting together to, to get the best thing out of these AIs. And really that's what AI is all about right now they call it prompt engineering. Right. Like that's what you're doing, you're trying to figure out what the best prompt is to get the results you're looking for. Right. And you mentioned Jasper basically. And I'm not a hundred percent sure, I've never looked at the code of Jasper. And you know, they, it's probably proprietary so they probably wouldn't, you know, confirm or deny. But I'm fairly certain Jasper's just using GPT3's technology and what they're doing is behind the scenes, they've engineered some additional prompts that get you the responses you're looking for. So when you're saying write me a blog post in the background, Jasper is adding additional information to your prompt to get the type of output that you're looking for. Right. So what we're becoming as we move more and more into these AI tools is we're becoming prompt engineers. Right? You want to learn how to be really, really good at prompting. And we're seeing it really a lot in the generative art. If you just go into like Dall E or stable diffusion and you say make me a picture of a monkey eating a cantaloupe on the moon, it'll be kind of a funky looking picture, right?

Yeah.

But if you say give me a monkey eating a cantaloupe on the moon in the style of.

Van Gogh, some.

Famous artist, Van Gogh. And I want it to be HD and detailed and colorful and in 4K and high detailed and look at Art Station for inspiration and you put all these additional prompts attached to it, you actually start to get really cool, colorful, way more creative images. Right. And so what, what we're doing right now is we're learning how to get really good at prompting. And so product prediction, here's where I think it's going to go in the future is you're going to see more and more tools pop up that are layers on top, right? So, so there's going to be a layer on top where I say, make me a monkey eating a cantaloupe on the moon. And behind the scenes it goes and adds all those extra prompts that I just listed off and then generates the art for you with my prompt plus the extra prompts.

So, like without even the user needing to do it, it's just kind of inherent. Like, okay, these are associated with their prompt. This is going to spice up the, you know, whatever the output is they're seeking, right?

And that's, and that's what I believe Jasper is doing right now with chat. GPT is there. You' giving it a very basic prompt, but it's just a layer on top of like a tool. It's either GPT3 or a tool like GPT3. So they're that layer on top. You put in a basic prompt, they go and add extra details to the prompt to get what you're looking for. And so what we're going to see, I believe, with generative art and generative text and those sort of things, is additional layers getting stacked on top. That's essentially how programming languages work, right? So in the beginning you had machine language which, you know, you have ones and zeros and everything's these ones and zeros. And then you have like another basic programming language that layers on top of it that, you know, it's got a little bit more readable language that tells what the ones and zeros to do. And then they stacked another layer on top of it that now you're looking at like a C or even more readable language. And then, and then you look at like these no code tools, which is like, you know, seven layers deep, where you're just dragging and dropping stuff. That dragon and dropping stuff is telling Python behind the scenes to write this code. That code is telling, you know, whatever, C to write this code. That C is telling the machine language to generate this. And, and so that's how coding works, is it's just layers upon layers upon layers upon layers of tools that are just telling the layer underneath it what to do, right? And so right now we're prompt engineering and we're basically. We're not the machine language, right? That's what's in the sort of the databases and the, you know.

These models that these things are trained on. Like that's the machine language that's behind the scenes, right? But we're working with the layer on top of it where we're translating it. Into like a more human language. But we really have to prompt the hell out of it to get what we want. I just think more layers are going to stack on top of it where next thing you know, you're going to just be dragging and dropping something and it's going to go through like four layers and spit out like a really, really cool image.

We're prompt engineers and that makes sense. And that means that's the new solution in the problem that we're faced with is being better prompters, better. And yeah, some of us are gonna be better at engineering these things than others. I already catch myself like overthinking things that I know when I get back to my like resourceful brain, not trying to just solve it within my own thinker up top is to like go to my computer, go to ChatGPT and just plug in that stupid question or whatever it might be. It could be anything and like it'll spit out that answer that's decent and maybe accurate up until 2021 at the time of this recording. And it's like now I don't have to wonder, wonder about it. It probably gives me prompts to think deeper on things. But yeah, with a better prompt or recipe or whatever thing that gives that inspiration to give the output better. I agree. I think it's a layered thing. Earlier, you know, you and I are chatting and I was like, man, this whole AI stacking concept, I mean, obviously I misspoke with machine learning stacking, but.

No, you misspeak. Everybody calls it AI. I mean, AI is the common language everybody's using right now. And it's just, you know, technically AI would be self learning and this stuff is very reliant on underlying models that people have trained.

Yeah, and it's all layers. I love the visual. I'm looking at this like the coding languages and it makes perfect, perfect sense with all these tools and the need to refine things to whatever the solution is or the output is that someone's after. And like I've made the response, the example, simple example of looking up a recipe. I looked up tomatillo salsa too, because I'm going to be making chicken enchilada dip for Christmas and also tapatillas also. So I was like, chad, gbt, I already know how to make this shit. But like, I just want to see what it's going to give me. And it gave me a cleaner recipe. It gave it to me like that I, I can generate it again if I didn't like that one. Maybe add some more Peppers in the next one or something. But, like, point. It gave me ideas like, oh, shit, I totally forgot I was going to have to buy. I wasn't going to buy that. I bought. Thank you, ChatGPT. Your OpenAI Google, you're out. But yeah, like, the prompting, like you said, and I put in an instant pot in one of them and it's like, it gave me very specific, refined to that in an outline. Like, I've used it for certain emails clarifying my emails.

I'll.

I'll just plug in an email I wrote and I'll say like, yeah, rewrite this in a professional way or maybe with a casual tone. And I haven't done all of this stuff and like, actually sent it, but I've done it to, like, see how it modifies my own writing. And it'll correct grammar. Like, I know Word Tune, it's an AI writing assistant. And like, it'll just refine your sentences. Well, technically, don't really need Wordtune too much. If you just use ChatGPT and prompt, it kind of does the same thing, different ui. And that's an example of probably, like stacking again. Yeah, it's just making the user interface easier. The experience, having a prompt.

Well, I use a Chrome extension called Quill Bot right now in my Chrome, and as I'm typing stuff, it suggests alternate ways to reword the things literally. It's like Grammarly, where as you're typing, it's literally making suggestions for me on Here's a better way to Word what you're saying. That one's called quillbot and it's just a Chrome extension that I've got installed. And it sounds like it does the same thing wordtune does. It's just, it does it everywhere. It does it as I'm writing tweets, it's doing it as I'm writing Facebook posts, it's doing it as I'm writing emails, all of that. Right? So, yeah, man, that's just another layer on top of a tool like Chat GPT. I don't know if all these are using chatgpt or not, but they're using a similar model and it's just stacking a layer and they're connecting into like that API or whatever. Right. But to go back to the question originally, where I kind of went off on a tangent of like, how does this replace Google? Right? You're just going to start going to the ChatGPT when you have questions on how to solve stuff, right? Like, like you said you might have used to go like chicken enchilada recipe into Google. Now you're going to go to chat GPT and say write me a chicken enchilada recipe, right? You might go to Google and like, like let's say I'm writing an article on the keto diet, right? You used to go to the to Google and say like just type in keto diet and find whatever blog post you could. Now instead, people are gonna go to ChatGPT and say, Write me an outline for a article about the keto diet. And then once they write the outline you're gonna say, all right, write me a paragraph on point number one, write me a paragraph on point number two. And then next thing you know, boom, boom, boom, boom, boom. You've got like a 7 paragraph, 1500 word article on the keto diet. And I still recommend going and tweaking it and altering it and fixing it up a little because it's not going to be perfect, but it's going to do, you know, 90% of that work for you. There is one sort of caveat to all of this. ChatGPT doesn't get everything right all of the time. I have found mistakes in there.

Oh yeah, it's definitely like it can flub on stuff so you better know your stuff and make sure to check it or rework with your own knowledge. I really see it as more of like, it's a, it's an inspiration tool. It's a thing that should prompt you even further to ask more questions.

Yeah, exactly. Because I was in there asking about how to do something on my website. I don't remember exactly the specifics and it was like, here's how you do it. Go to this URL and type this in. And it was giving me step by step instructions. Only problem was the first URL that it suggested didn't exist.

Interesting. Yeah, there's, there's, there's some drawbacks but again, it's going to get better. This is just like, it's just now getting all of this data. I saw someone post on Twitter about how they use ChatGPT to mine it to essentially create a script for a movie. And then it was going to do everything with AI, I think it was. Zenica shared this in his newsletter. You probably read it. Yeah, it's essentially it was like going to chat GPT asking them to write this whole like script for a, a feature film or a short or whatever. But it's like, use the hero's journey, use all that. Like you can deck it out and it'll outline it. And then like you said, started going into it like, okay, describe the appearance of all the characters and one by one, okay, now you have a whole description of those characters. Now you take those character descriptions, plug them into like an avatar creator or a character creator, you know, make some generative video. Now using that prompt where ChatGPT gave you all that. I'm sure there's like maybe a smoother way to do that now, but it just shows you how you can, you can just prompt and then essentially build from there into this whole elaborate thing. You probably never.

Well, that's the way. So you know how the ChatGPT recently added the ability to go back through old chats? Well, the way they want you to do it is to kind of have different conversations. Because the way ChatGPT is working is it's actually using your existing conversation to contextualize your next question. Right. So like, if I'm sitting there asking about the Keto Diet, and then all of a sudden I change the topic and start asking it, you know, questions about how do I write code in Python or something, it's going to get a little confused because the context of the conversation to this point has been about the Keto Diet. Makes sense. And so the reason they added that bar on the left to start new chats is that so you can start a brand new chat with, with new context.

Got it.

Right. So like, if you want to write like an article using ChatGPT, that's exactly how you do it. You'd ask it to start by writing an outline. And then you'd go, okay, bullet number one, from your outline above, write a paragraph above it, bullet number two, and then the next thing could be write a paragraph about bullet number two, write a paragraph about bullet number three, and because it knows the context of the conversation above it, it knows how to write the next thing.

Got it? See, I didn't know that, but that makes perfect sense. I saw that feature and now that you describe it. Yes, it's going to learn from our input and then we're going to learn it, learn from it if we choose to. Or we're going to say regenerate that. I don't like that, you know, and it's going to spit out something else. And then we go deeper on that thing. Damn, I'm going to. I can't do. Mind blowing. And I'm sure most of us have used it or at least heard about it. If you don't have any, just go sign up and start playing around, like, spend 30 minutes and blow your mind. So. All right, I'm going to put you on the spot. I want to get into tools, and I know we talked a lot about tools already, but I want to hear your top 10 tool stack. These AI tools doesn't. I know you're rolling your eyes because it's a lot, but you're going to do it.

10 is a lot.

I know. Well, good thing we have future tools. AI. If you want to cheat, you can go to your own website and start just using the search features to find tools that you know you use often. The browser tool, that can count for one. I already downloaded it, by the way. Quill. Quill Bot. Quill Bot, yeah. That's a good one.

Yeah, but I. I like it.

Fire off the tools. No order needed.

All right, well, I'm going to cheat because I'm going to say ChatGPT is one of them.

That's fine. That's fair. I'm going to write these down. So Chat GPT.

Yeah, so Chat GPT. Now, when it comes to, like, I've done a lot with generative art, right? My favorite generative art platform is Midjourney. That's the one that you actually build out your generative art directly inside of Discord. So you actually type your prompt into Discord and it'll spit out images for you. And that one, they'll let you do like, I don't know, 10 images for free or something like that. And then you could upgrade. And for 10 bucks a month, I think you can do up to 200 images. I blew way past the 200 images, and now I'm on their $30 a month plan.

Where is that when you're making all your selfies?

No, that was stable diffusion. That's a completely different model. It's a different training tool. So what I was making the. So that would probably be my third tool would be stable diffusion. Now, stable diffusion is. That's one where it's really easy for anybody to train their own stuff on, like mid Journey can kind of train it on your own image. Like, I can throw one image of myself in there and say, you know, make an image like this one, but in the style of X and it'll try to make it look like my face on that style. But it's usually not very good. It usually doesn't look too close.

The.

With stable diffusion. Stable diffusion, you can actually train it on whatever model you want, right? If, if you want to upload a whole bunch of like, let's say you're an artist. I know your sister's an artist, right? Let's say she wanted to train a model on her style of art. She can upload every picture she's ever painted into this and say, like, you know, this is the. This is the Sarah model or whatever, right? She could upload every single image in there, call it the Sarah model, and then start saying, all right, now generate a dragon in the style of the Sarah model. And it will generate a picture of a dragon in the style of her artwork because she trained it on all of her art. That's the stuff that all the people are getting outraged about in the generative art, which maybe we'll get into, maybe we won't. That's the type of thing that everybody's, you know, that's the tool that people are having shit fits about right now. But with Stable Diffusion, you can actually install it on your own computer and generate. You have to have a decent gpu, right? Like, I've got the Nvidia, whatever, a gaming graphics card, so it can handle it. But not every computer can handle running stable diffusion on their computer. There are online ways to do it where you use, like, remote graph GPUs, right? But that's one where you can actually train the model on whatever you want. And I uploaded about 25 pictures of my face. You want to upload pictures of your face from your waist up and your whole body. And so I uploaded, like, 25 images. It's trained on what I look like now.

Were you in the mirror doing something?

No, I just. I just took a bunch of pictures off of Facebook and trained it on a bunch of pictures that I had already. So, like, I trained that one, and now that model's trained. Now I can spit out as many images of my likeness as I want. And there's a bunch of tools out there that have basically monetized stable diffusion. So I'm sure you've seen tools like on. On the iPhone where you can upload a bunch of pictures from your iPhone and it'll spit out, like an avatar in the style of a cartoon or a superhero, whatever. All they're using. All they're doing is you're using stable diffusion, right?

So what's the name of one of those that have. You used one? Let's. Let's mark one off the list, just so we help you.

Yeah, so there's AI Time Machine, Avatar, AI Hero Pack, Magic, AI Avatars, Magic Pick, Makelogo AI and Prisma. Prisma is probably the most popular one, right? Now that's the one that's on the iPhone where I think for like you upload 25 pictures and then for 20 bucks they'll give you like 100 images generated. But all that's doing is using stable diffusion behind the scene and giving you the images back.

Got it.

So that's stable diffusion. And then there's Dall E, which is the other generative image one which was kind of the first one to rise to popularity. But in comparison to Stable diffusion and Mid Journey, most of the images kind of suck that come out of Dolly.

So you would choose. Yeah, and I'm trying to. Yeah, you're down to stable diffusion and mid journey. Okay. So in terms of the art, generative art, there's different uses for those, of course. And then Prisma is another art related type thing. Okay, how about Generative Gen er, A tive. There you go. Video. Have you used any of those?

So I used a tool called movio. Um, that was one that I actually put a video on Twitter. You might have even seen it where it was like my animated avatar and I was talking and I was like, hi, my name is eflow that was using movio, which you can actually upload like an image of your face or like one of these headshots that you make from Stable diffusion. And then you type in text and it will actually speak out the text and animate your mouth moving to it. Or you can do like a voiceover on it and it will match the mouth to like the voiceover that you record into the system. That one's called movio M O V I O. Yep. And then there's another one called Synthesis or Synthea.

Yeah, Synthesia.

Yeah, Synthesia. That's almost an exact copy of movio. Like you look at the two sites side by side and they're pretty close.

Oh, really? Yeah. And I love how you have like a little description. So that's description is AI, text to video, add voice and speaking animation to avatars. Yeah, yeah, gotcha. Cool.

So I'll tell you another one that's like blew my mind when I started playing with it.

What is it?

It is called Voice AI and it's free to use. And it's one where I can talk into my microphone and it will translate my voice. It's like a voice modulator. It will modulate my voice into like anybody I want. I can sound like Elon, I can sound like Morgan Freeman. I made a recording of me and the voice of Rick and then a version of me and the Morty and I had made Rick and Morty have a conversation with each other. I've done some recordings in the voice of Joe Biden and they are frickin good. Like, if you listen closely, you can hear a little bit of like digitization to the audio, but it's, it's hard if you're not paying attention. And if you underlay some music under it, you definitely can't tell. Wow, you can actually do it in real time. I've actually logged, I've actually opened up like DaVinci Resolve and just recorded my voice talking and had like a whole 10 minute conversation, listen back. And I sounded like Morgan Freeman, right? What I did was I actually generated a story. I generated a story about a crypto buffalo. I said, write me a story about a crypto buffalo. And it wrote this story about this buffalo who lived in the plains and how he lived a lonely life until he discovered a computer. And then we got on the computer, he discovered the blockchain and got involved in crypto. And then he started teaching others about crypto and. And then the next thing he know, everybody started calling him crypto buffalo. And he became this hero among buffaloes or something like that. And I was like, that's a really cool story. I want it read in the voice of Morgan Freeman. So I jumped into voice AI, opened up my DaVinci resolve, recorded the entire like story that it wrote out in the voice of Morgan Freeman telling the story of the crypto buffalo. Right. And both tools were completely free. I did it. I mean, even DaVinci Resolve was free that whole process. 100% free.

That's a good point. And DaVinci Resolve, if you all aren't using that, it's amazing if you want to get into video and audio editing for yourself because, yeah, I've been doing the same. So to help you out as well. And I know we mentioned it, I think at the top of this episode was Time Bolt. I want to give a shout out again there for a podcast AI machine learning type tool. That's what I'm now using at the time of editing after this podcast, obviously. But yeah, it's a cool tool that will. It actually has a downloadable interface that you put in your computer and you upload the tracks into it does video as well. It's not just audio and it will chop it up, it will do all the jump cuts and all this crazy stuff and it actually allows you to kind of see the timeline. You can do a lot of your editing or dare I say all of Your editing. If you want to get something out quick, right there in Time Bolt. And that's. That's one of your things. I figured I would help you out and also add that to the top 10 list, because that blew me away. And I tested the other one. You know, the other. There's a wise cut. It's a. It's more of an online user interface. And like I was mentioning earlier, it's just a lot more simplified. It did a great job. Um, so I think those are both great tools, but I like Time Bolt more just for the customization of it.

Yeah, I haven't actually played with either of the tools a whole lot yet, so don't know the pros and cons of each too much yet, but those are on my list to dig into a little more.

I gave them both a free run just to see how to use it. So that's my quick feedback there. Yeah. Hey, what's the tool? And I'll help you out again, Matt, because you made a video on this, on the Slide Deck tool. Would you add that to the top 10 list?

Yeah, that's called tome T O M e Tome App. So that's T O M E app. And what Tome does is you plug in a prompt that's basically anything. It's, you know, it's another layer on top of the OpenAI stuff. Right. If you watch my video, you can see that the image generation that it's grabbing the slides from. The images on the slides is actually the. The DALL E2 images. Right. The text generation is also, I believe, using, you know, the GPT3. So it's. It's using all of the open AI sort of source material. And what it's doing is you can type in a prompt that's just like anything. The example I used in my video was write me a presentation on the Keto Diet, right? And what it did was it made a 10 slide presentation of like all the benefit. Each slide had a different benefit of the Keto Diet, and each slide had an image that kind of sort of represented what they were talking about on that slide. And then if you don't like the image that it does, you can tell it to regenerate a new image. It'll do the AI and make a new image. And it makes an entire slide deck for you from scratch by just a simple prompt like, make me a presentation about the Keto Diet. That's another example of just another layer on top of existing tools like GPT3 and Dolly too.

Bitching. I love It. Yeah, man, that's. I watched the video and definitely go seek out some of Matt's videos. I'm sure you'll be putting out more because these tools as they go live, I think one of your goals is to basically do a video walkthrough. And I always like the way you teach this stuff, so it gives me more ideas on how to apply it too. Hey, so the music app. Let me help you out with another one.

Sound Draw.

Yeah, Sound Draw. Okay, so yeah.

S O N D E R A W. Yeah, and that's probably the one you used because that's the only one that I know of that's publicly usable at the moment.

Yeah, because I tried use. Net and that's another open AI and I don't think it's public yet.

No, you can. You can listen to some examples of stuff that it generated, but you can't really generate yourself that I know of yet.

Right.

But yeah, with Sound Draw, that one's really cool because that's the one where you can tell it the length you want it, the tone you want it. Like, I want to. I'm making a scary movie, so I want it to be a scary tone. Or, you know, I want it to be high energy for like a gaming video that I'm making. But then you can actually get into like the. The audio that it creates and say, all right, at 30 seconds, I want it to ramp up this. The pace at, you know, a minute and a half. I want the beat to drop right here. Right. And you can literally sort of customize the music.

Turn the knob up and down.

Yeah, you basically turn knobs throughout the video. The audio of, like, how you want it to line up with your video. So you could be making a gaming video where you just like walking around and all of a sudden you shoot at something and the beat drops and you can make that exact song. And it'll be a completely unique song because it was generated with AI and it's royalty free because it's not created by a human. I mean, you have to have a Sound Draw account to be able to use it, you know, for commercial purposes and stuff. But it's totally royalty free. Nobody else is going to have that song in their video.

Dude, man. Yes. That's. So I haven't actually, like, downloaded it, but yeah, I started twisting knobs and I was thinking about the application and yeah, it's just twisted knobs exactly like that. You know, you can imagine what that looks like. But it's simple. That's. I think the biggest thing, I mean, man, I actually tried to use that website for the intro because I made a new intro for this podcast and it took me forever. I actually ended up going to Motion Array. That's not an AI type tool, but it's an awesome resource for tons of music and stock footage and pictures and whatever else templates, DaVinci resolve type stuff too. But yeah, it's like music is such a pain in the ass. And that's, that's a good one. If you have like a length, like a 10 minute, 15 minute long video or whatever, you're shooting three minute long. But yeah, you put the time length, all that stuff, the emotions rise and falls as you go through. It's really cool. How about a productivity tool for rounding out the top 10? I have one in mind, which I can say. Or if you have a better one.

Um, let me actually look. I'm going to go over to FutureTools IO, not shamelessly plugging it at all. And I'm going to click on the productivity tab here.

I mean, that's what I'm looking at too.

So I mean, there's a lot of like. All right, so a couple really, really cool ones that I've played with. The one you might be thinking of is Otter. Otter is really freaking cool.

It's not, but I use Otter. I pay for it. It's great. It's usually the tool I use in the spa when I'm thinking about like some brilliant idea that I feel like is brilliant. And then I feel like I need to have a transcription. So it's a virtual note taker. It works great.

Yeah. So the one that I think is really cool is there's a tool called Explain Paper and it's free, I believe. Let me just double check on that. Yep, it's free to use. And what Explain Paper does is you can take like, you know, when we were doing a lot of blockchain stuff, we were reading a lot of like white papers for games and tools and Defi platforms. And the white papers can be pretty complex and confusing. To understand what you do with Explain Paper, you just drop that text file in there and it'll be like, here's basically what this paper is all about. And it'll sort of synopsize it for you and explain what the white paper was all about. Right. It's literally that you take complex documents and it simplifies them for you.

Run it by your lawyer. If it's a legal document, maybe you could attempt it, but I don't know How. But like, that is awesome. I was eyeballing that and I've heard of Explain Paper from some Twitter thread. Um, and yeah, that. That's brilliant. And I think I'm going to give that a run. I have some ideas, some use cases. I want to run that by some contracts. So that's rad. The one I was going to say. And maybe this is a bonus one. It's actually one you mentioned to me earlier. Maybe. I think it was off recording, though. That's why I wanted to bring it up. Is the GLASP YouTube summarizer. Is that the one that. That's the one. Okay, so basically it runs a. Maybe a long YouTube video thing through its GPT system, whatever, and then it summarizes it for you.

Yeah, so that one actually what it does is it's a chrome extension and it goes and it scrapes the transcripts from a YouTube video and then it just throws them into your ChatGPT. Right? So like, if you have a login to ChatGPT, it's just like a simplifier. It'll go, okay, grab this transcript, throw it to chatgpt, tell chatgpt to summarize it for me. And then in your own chatgpt, you'll actually see it summarize for you. Right. So that's like a shortcut tool to go and grab YouTube videos and then pull them into your own chatgpt and get them summarized. So it just. It saves a couple steps of you copying and pasting, you know, the transcripts yourself is essentially what that one does. Another tool that you and I have used a lot over the years is Notion. Notion actually has AI built into it. It's basically got like its own version of a chatgpt in there where you could. I actually haven't played with it a whole lot yet, but you can go inside of Notion and in a cell, tell it to like describe the thing in the cell to the left and it'll write up a description for you with its own sort of chat thing. And I actually think it might be. Because Notion, I'm not sure it's. Yeah.

Really? Yeah, because Notion has its own. I mean, you could build out websites on there, all sorts of different things that are public. I wouldn't call it a no code solution. I mean, it is, I guess, technically more database, but wikis and notes and stuff like that that you can share publicly too. Yeah, that's a prompting would take it to another level. This is insane. Matt. I'm not going to sleep again. Damn. It AI assistant for video calls relayed. I don't know what that's all about. I think the big, like, I really want to see recipes and concoctions from you, like from you as a marketer, as you Matt, that people are hearing that people on this show will go and want to go get. How. How would Matt assemble something to do like whatever that thing is, produce this video, produce a slideshow that can then turn into a voiceover video that you're actually typing into text or maybe talk it through through a different voice. Maybe you're working in a different niche. I don't know what the hell, but like applications for all this stuff, I don't know. I feel like there's some cool thing there that you can add to the site that people would nerve.

I've already got an idea that I want to test next with that tome tool where I'm going to have it write an entire video for me, basically make all the slides, the images, the text, and then basically copy and paste all the text that it generates for me. Put it into a text to speech generator, have the text to speech generator actually voice over everything that's going on in the slides and then, you know, just combine the two and then you have a basically a made for you video that, you know, all you have to do is sort of layer the video that that tool spits out and the audio that the other tool spits out. And now you've got like a video you can publish on YouTube that if somebody wants to make a YouTube channel but they don't want to be on camera all the time or whatever, there's a process, you know, so there's a.

Creator in all of us. All we need to do is think about how to attach these things together, be a better prompt engineer. Right, There you go. I think that's really it. Prompt engineering is going to be a thing now.

Oh, it's a thing. I mean that's a. That's a very popular term that you'll see among the sort of AI generative face. Right. It's prompt engineering is what everybody's doing. They're trying to figure out how to get their prompts to do better things than other people have prompted so far.

Well, I want to, I want to wrap this up. And now that we have so many freaking tools, my mind is like, I want to start jumping down the rabbit holes. I'm sure most people already have been. Have they been as they've been listening to us? To wrap this up, are there any final thoughts or maybe predictions you want to put out there for 20, 23 and, and in the shorter term because we're going to chat more so you don't need to do some lofty thing.

Yeah, I mean, I feel like we've kind of thrown out a lot of the predictions. I think what we're going to see is we're going to see just a crap ton of SaaS products pop up because of both. Like, remember when we talked to Mark Joyner, right, He had that, what did he call it, convergence or.

I believe so, yeah. Like spot on with like everything in that book.

So, yeah, like what we're seeing right now is the no code tools, which makes it really easy to make websites, really easy to make SaaS software, really easy to pretty much make anything you can imagine online. There's no code tools that can build it. But then we're also seeing these AI tools that do a lot of the writing and the art for you and a lot of the sort of creative processes we're seeing that really on the rise. So we're seeing the convergence of those things right now, like no code and AI. And because of that we're just going to see this prolification. I like that word. We're going to see just like a shit ton of SaaS products pop up, right? We're going to see a ton of these SaaS products pop up everywhere because it's easy for anybody to go build a tool now, use a no code tool, connect it to an API of an existing AI tool and boom, you've got a software that makes people's lives easier. So we're going to see that stuff pop up like crazy. And you know, the marketers are going to kill it, right? Like Gary Vee says, marketers ruin everything. Marketers are going to go spit out shitty SaaS software, but there's also going to be some really, really cool SaaS software to come out of it. We're going to see people probably using ChatGPT a lot more than Google. Like, here's the other thing. Google is working on their own version of like what ChatGPT is right now, of course, but they've sort of held back on releasing it due to sort of ethical concerns. Right? They're still, they have some like, they were worried about putting it out there, but. But now that ChatGPT is really growing really fast. I mean, it's the fastest growing software in like the history of software. It got to 1 million users in like a one week time and it took like Facebook, you know, three months or something like that to get there. I'm pulling these numbers out of my butt, but it's like nothing here.

I'm gonna be the Jamie and look.

It up for a second. Cause I want to get stats. But keep going, keep going.

Yeah, ChatGPT was like basically the fastest growing software like ever. And like as far as like active users on it. And so we're going to see people going there instead of Google and you know that's making Google poop their pants a little bit. So they're going to have to speed up the, the release of their version. So what we'll probably see, and I don't know if it's going to happen in 2023 or not, but what we are going to see eventually is a lot of this AI stuff built right into tools like Google, where you're going to start searching and if Google can answer it for, for you, Google's not going to send you off of their site to somewhere else. Google's just going to answer it for you.

Right?

So we're going to see that happen. It's going to fuck the SEO industry, like hands down, it's going to, right? So that is something else that we're going to see come from this. You know, there's going to be a, it's going to be a very hot button topic. We're already seeing the topic of ethics bounce around, especially with the generative art, because you know, there's, it's coming into question where these models were trained on, and these models were trained on publicly available art, which now people are able to create art in the style of somebody else. And the person who originally created that art style is getting nothing out of it. And so that's affecting their livelihoods, right? Like imagine if anybody can go and create a Disney movie now because they have that technology. Disney's not going to be happy about that. Right? So we're going to see the sort of ethical debates. We're going to see some companies trying to squash it down because it hurts their business model. But the genie's out of the bottle, right? Like, oh yeah, this shit exists. Nobody's taking it away. If you try to take it away, it's just going to make people figure out ways to circumvent whatever restrictions get put on it.

Right?

There was already a thing with stable diffusion where the creators of stable diffusion removed a whole bunch of not safe for work content from the models because they didn't want people, you know, going and taking Angelina's Jolie's head and putting it on somebody, you know, sticking it in somebody else's butt or something. Right?

Like they, it's been done, I'm sure. And they're now blocking all of that. Yes.

Yeah. So like, you know, we were seeing the, like we saw stable diffusion step backwards and say, you know, hey, we don't want you making this kind of content with our platform, right? And there was a big uproar over that of people going, we should be able to be allowed to do whatever.

We want with this, like decentralized machine learning.

So like, you know, it's, it's going to be a divisive topic. Like, who would have thought that like these cool AI tools that you and I are nerding out over would be something that is going to probably get legislation involved and governments and all that. I thought I was getting away from that when I stopped focusing on crypto as much, but there's just as much of that shit going on in this space right now because it's such uncharted territory.

It's money and porn always seems to be on the leading edge of a lot of technology. So. And yeah, so user base around the globe. So I'm sure it will be used for those types of things and nefarious things. I mean that's always going to happen with tech. It like new, new bottle, genie bottles pop open. All these other places using different tech.

And oh yeah, there's, I mean, you know, the thing about stable diffusion is anybody can train their own model on anything they want, upload it. So if the creators of stable diffusion said, hey, our new models don't have not safe for work content on it. Anybody that's using stable diffusion is like, cool, we're not going to use that model. We're going to go create our own that does. And there's already like this group called unstable diffusion where they're creating their own models with nothing excluded inside of the models, you know, so like it's just.

You know, the database is there. So yeah, we're seeing all the genies pop up and they're like, get back in there. Like you can't see that or you can't create that, which they probably have some good reason for it in some cases, maybe not. Who knows? I'm not the one to say. But the fact is, yeah, it's such a wild west kind of space right now with all the data and it's just the beginning. This is, I think Zenica even said it in that newsletter I mentioned is like, it's the end of the world as we know it.

But I feel fine.

I do feel fine. I mean, I really. I think it's interesting, and it leads us to asking more questions. And I think, like you said earlier, we're actively getting smarter, and I think that's the truth. And it's. I know. It's like a reshaping of our habits and our. You know, the ways that we solve problems will be right now, like this. This time, people are engineering things in ways that we never have thought of before, and I think it will upgrade our minds in the way that we solve problems, and it's just going to introduce different opportunities and then problems to solve and things that. So I do feel fine. That's why you can.

You can. You know, ultimately, you can either fight against our new AI overlords or you can work with them. And I, for one, choose to work with them.

Let's do it. All right, so let's wrap a bow around this one. This has been great, man. I want to do more of this. So you coming back? Sure. All right. Not as a host, but, you know, no commitments here. I'll make sure that.

Dude, I'm down to do this, like, once a month because, like, you know me, I go down rabbit holes and I stay there for a long time, and then I find out how deep the rabbit hole can go, and then the next thing you know, I'm on. I'm in Australia.

That's not a bad place to hang out. There's a lot of friends down there, too, so I'm sure you'll make a lot more on the way in this AI universe, too. What's that website where you could go to and just see, like, new humans all the time that don't exist? It's like, this is not a human or something.

This person does not exist.

Dot com.

Yeah, Dot com.

Yeah. If you just refresh the page, you just.

It'll just be a new picture. Yep.

Someone that is not actually real, but looks pretty damn real. Yeah.

And I'm. I'm pretty sure, you know, Twitter bots and stuff that want to look like real people that aren't looking like real people are probably scraping sites like that right now.

Dude, it's. It's wild. The world is crazy. But you know what? Let's lean into it and have fun. All right, so. All right, man. We'll talk soon. Go check out futuretools IO, Sign up to the email list. Go use that book market legit. You will go down rabbit holes and solve all of your problems. No, it's not financial advice, doctor advice, or any of that kind of advice.

Medical advice, but it cured my hemorrhoids.

Maybe ChatGPT will lead you that direction. Who knows? Anything is possible. So we'll talk to you soon. All right? All right. That's what I got for you today. And remember, it's not all about living to work. It's all about working to live if you need to do the work. But you know, work can be fun and it should be fueling for your life. So if you enjoyed what you just heard, if you got some nuggets of wisdom that you want to share or you're just noodling on right now, please go tell others. The way that people find this show and how you can help others get their aha moments is through word of mouth. So if it is telling your friend, telling your family, send an email to your list, writing a review or whatever it might be. Everything helps. So thank you so much for listening to the Hustle and Flowchart podcast and I will see you next time.