콘텐츠 제작 및 관리를 위한 일상적인 AI 도구

### 일상적인 생산성 및 콘텐츠 제작을 위한 나의 필수 AI 도구
* 🤖 저는 매일 다양한 AI 도구를 탐색하지만, 그중에서도 **콘텐츠 제작과 생산성 향상을 위해 꾸준히 사용하는 몇 가지 도구**가 눈에 띕니다.
* 💬 제가 가장 먼저 사용하는 도구는 **Claude**로, 대화와 프로젝트 관리를 위해 항상 열어두고 **기사 정리 및 YouTube 제목 생성** 등의 작업에 이 도구를 활용하고 있습니다.
* 📝 예를 들어, 금요일 뉴스 동영상을 준비할 때 Claude는 **지저분한 URL을 깔끔한 형식**으로 정리하여 YouTube 설명에 쉽게 복사할 수 있도록 합니다.
* 🔍 최신 정보가 필요할 때는 무작위 질문에 대한 빠른 답변을 찾기 위해 **기본 검색 엔진으로 설정한 Perplexity**를 이용합니다.
* 📚 또 다른 필수 도구는 내 관심사에 따라 뉴스 기사를 집계하고 시간이 지남에 따라 관련 콘텐츠를 표시하도록 AI를 학습시킬 수 있는 **Feedly**입니다.
* 🗂 ️ 내 마음은 기사 및 동영상 저장을 위해 Evernote를 대체해 **AI 태그 지정과 요약**을 활용해 정보를 쉽게 찾고 정리할 수 있도록 도와줍니다.
* 🖼 이미지 생성의 경우, 저는 주로 **레오나르도를 사용하여 AI가 생성한 썸네일**을 만들고, **Stable Diffusion을 사용하면 이미지의 얼굴을 교체**하여 개인화된 느낌을 줄 수 있습니다.
* 🎶 마지막으로 동영상에 사용할 맞춤 음악을 생성할 때는 **Sunno**를, 프레젠테이션 필사에는 **Otter**를 사용하여 워크플로우를 더욱 효율적이고 즐겁게 만듭니다.

Claude는 저자가 가장 많이 사용하는 AI 도구로, 일상 업무를 간소화하고 생산성을 향상시킵니다.

Not a day goes by that I don't get asked the question, What AI tools do you actually use? Well, first of all, I'm probably the worst person to ask that question because I literally test them all. When somebody tells me about a new AI tool, I go and play around with it and test it. So, I literally use all of them, but there's just a handful that I find myself using pretty much every single day, and that's what this video is going to be. I'm going to show you the tools that I use that leverage AI, that I'm pretty much using every single day. Number one for me is Claude. I'm in Claude constantly, literally. This is always open. I always have a Claude tab open on my dashboard. If I come over to my sidebar here and click on View All under Recents, you can see that I'm pretty much constantly having conversations with Claude all the time, every single day. Last message: 1 hour ago. Previous to that: 17 hours ago, 2 days ago, 2 days ago, 2 days ago, 3 days ago, 3 days.

ago 3 days ago 3 days ago 3 days ago and so on. I'm using it constantly. Most of what I'm doing these days is leveraging the projects feature. If you're familiar with chat GPT, they have a functionality called gpts, where you can give it custom instructions and add additional information to it, and then each custom GPT will sort of be more focused on that single task you're trying to do. That's kind of what projects do. So, I have 10 different projects in here, and I pretty much use all of these projects. I'm going to make a separate video that goes through each of these projects, how they're set up, how I use them, and give you some other ideas for projects in your own business. But here's a quick overview. So, I've got an article cleanup project here, and this is whenever I make my Friday news videos. I give it a long list of all of the websites I'm going to share, and then it sort of organizes and cleans up all of those articles for me.

it cleans up the URL. It cleans up the title so that way I can copy and paste into the description. In fact, here let me just show you. This is my most recent Friday news video. I posted in a giant list of URLs that I was going to be talking about. You can see how messy and ugly that looks. And what it did was it spit out a clean version. Open AI, Strawberry AI, and then the URL Nvidia, Open AI funding, and then the URL. So it just cleans up all of the URLs. And then I can copy and paste what this outputs into my YouTube description. So all the links of everything I talked about are there and clean and easy to read. I also have a YouTube title generator. I feed it a video concept and get ideas for potential titles. Literally, what I'll typically do is I'll record a video, get that video transcribed, throw the transcription into the YouTube title generator, and it will come up with a bunch of titles for me. I was making a video about Runway Gen 3. I uploaded my

transcript. You can see some of the titles it suggested: This AI makes insane videos in seconds, Runway gen 3, The future of video is here, etc. I have an ad integration writer, so when I do sponsorships on my YouTube videos, a lot of times the script that I'm reading to explain the sponsorship was written by this ad integration writer. I get a brief, basically a summary of what the sponsor wants me to talk about. I plug the brief into the integration writer. Claude writes the integration for me. Now, pretty much all of this stuff, I kind of clean it up. Sometimes it's not perfectly to my voice, so I will reword it a little bit when I'm actually reciting it on video. But for the most part, it gives me a really clean, almost ready-to-use rough draft. This second brain project is actually a coding project that I've been working on. That'll be a future video where I'll dive into it. I have a YouTube consultant where I uploaded a whole bunch of blog posts and transcripts from

videos and things like that that talk about how to be better at growing a YouTube channel. I fed that all into the project, and now I sit around from time to time and have conversations about how to grow on YouTube. I have my flux image prompt optimizer. This is actually a project I made on camera in one of my previous flux videos. I plug in a real basic prompt; it gives me a much better, more optimized prompt for flux, an audio journal app. Once again, this is a coding project I'm working on, which will be in a future video. I've got my new summarizer. You can input a PDF article or transcript and get what I need to know out of it. So, for example, I uploaded an article about an AI-powered solar energy breakthrough. It gave me a summary, it gave me a breakdown of what makes it novel and interesting, and then it gave me some analogies that I can use to better explain on video what that news article is about. A lot of times when I'm explaining things on videos, I'll run

through a CLA first to try to find some good analogies so that I can share it and teach it in a way that hopefully will be easy to understand. And then I've got two shorts writers. One is for tutorial video, so if I'm going to write a short like 60 seconds or less that explains how to do something, I can plug in the longer form tutorial or type up my explanation of how to do something, and it will rewrite it into a short that is 60 seconds or less. Here's one where I uploaded a tutorial on how to turn screenshots into functioning apps with Claud, and then we can see here it wrote a 60-second script on how to do that process. And then I have my shorts WR, which is probably the one I use the most. I uploaded a ton of different shorts transcripts that I really, really liked, and then in the custom instructions, I told it to create video scripts that will be 1 minute or less in the style of the scripts in the project knowledge.

basically, these are all scripts that I really, really like. I give it some information about a video that I want to produce, and then it writes me a script in the style of all of my favorite videos over here so that I can then turn around and make a short. And I use this one a lot, so Claude, my number one most-used AI tool. In fact, like I mentioned, I'm going to do a full breakdown on all the ways I use Claude. It's pretty much fully replaced ChatGPT for me. I have nothing against ChatGPT; I just like the interface and the results I get from Claude a little bit better right now. Now, I'm not super loyal to Claude. If ChatGPT improves at some point, I might switch back. I have no loyalty to Claude or ChatGPT or any of these platforms. Whatever one gives me the best result is the one I'm going to use right now. It's Claude. Claude does have one major downfall, and that is that it's not able to search the internet for me. So when I

have a question that I want answered that requires more up-to-date data where it needs to search the internet well in those scenarios I use perplexity. I do have a perplexity PR membership. I probably should have also mentioned that my cloud membership is also a pro membership, but I use perplexity constantly. In fact, I even changed my Google Chrome bar up here so that if I search something, it searches perplexity by default. So if I was to come up to Chrome up here, you can see it's got the perplexity logo when the URL bar is blank, and I can literally ask, you know, how far is Mars from the Earth right now. And when I type into the search bar, it actually opens perplexity by default instead of Chrome. I get a quick answer here, and it gives me everything I need to know. Now, if you want to add this to your own Chrome, pretty simple. Go up to the three dots up here, come down to your settings, over on the left sidebar, you've got search engine, and then under search engine, you can click this

little change button and select perplexity. Now, if perplexity isn't on this list, you might need to click this little button to manage search engines and site search down under inactive shortcuts down here. Open this up and search until you find perplexity. I've already added it, but when you find perplexity, go ahead and click activate. And then up here, you just click the three dots, and then you would set perplexity as your default. That will change it so it automatically searches perplexity whenever you type something into your Chrome bar. Now, one thing that perplexity has really enabled me to do is whenever I have any random thought, I open up perplexity. And I actually use the mobile app for perplexity quite a bit as well. But anytime I have any sort of random thought, I type it into perplexity. It's actually gotten to a point where it's pretty ridiculous how often I just have a random thought and then go ask perplexity. So, if I look at some of my search history, I was asking

questions about the otter app. I randomly thought I remember when magnets used to screw up computers. Does it still do that? Do magnets have any impact on modern laptops or solid-state drives? I heard rumors that Stan Kony is building a new stadium in San Diego, so I asked. We were watching a Padre game, and my wife asked, Does Min Machado have kids? I was like, I don't know. Let me ask perplexity. So I asked perplexity. I started taking Elderberry for immunity. I asked what of the side effects, and then I started going down this side effects rabbit hole. I ran into an issue with YouTube shorts, so I asked perplexity. I ran into an issue with my mouse, so I asked perplexity. And as you can see, whenever I have a random question, I pop it into perplexity and get an answer. I use it constantly. So when I need AI to sort of do the work for me and help me write a rough draft for a script or clean up.

articles or things like that. CLA is my go-to when I have random questions that I just want that question answered. Perplexity is my go-to next up for me is this tool Feedle. Now Feedle for the most part is a pretty simple RSS feed reader. If you have blogs or news websites that you read constantly, you can come to this site, add the RSS feed to the site. Basically, you just plug in the URL here and it will find the RSS feed for that website. Whenever a new post gets published to that website, it updates it in Feedle. But Feedle recently added a new feature called Leo. And if I look on my left sidebar here, you can see a little AI box that says create AI feed. And in here, it's basically like an advanced version of Google Alerts. So instead of plugging in websites or blogs that you want to make sure you see every post from, you put in topics. So I can put in artificial intelligence, machine learning, diffusion models, things like that that I pay a lot of attention to.

and anytime that stuff pops up, it shows it in my little AI web alerts tab here. Now, I've already reviewed my AI web alerts tab for the day, and I'm all caught up. Let me uncheck this so it'll show you everything that's popped up. Now, not everything it shows is AI-related; sometimes, it'll just be like the article mentioned AI somewhere in it. But the majority of the news isn't related to AI. But we can see like Ant International and Doc are improving financial inclusion in Brazil with aid-driven credit tech, Before Mid Journey, there was Night Cafe, and it's still kicking, Sign or veto: what's next for California's AI disaster bill SB147, and these are all websites that I wouldn't normally just be subscribed to. Like, I'm not subscribed to the Fintech Times, I'm not subscribed to Coin Spectre, but because I have the keywords related to AI in there, it surfaces those posts even though I'm not specifically subscribed to that blog. Now, how is that different than a Google alert? Well, over

time. I can actually train this feed to show me more or less of what I want to see. For example, Oasis ticketing chaos prompts probes into Dynamic pricing, not super relevant to what I'm looking for. So, I can actually click this button that says less like this, and over time, it will train your feed to show you less of what you don't want and more of what you do want. So, you can constantly train this Feedly AI bot to just show you more and more of the news articles that you actually want to see. Anybody who asks me, How do you stay so on top of the AI news? How do you have your finger on the pulse of the latest tools and products and news and launches and stuff? This is it. Between Feedly that uncovers a whole bunch of news for me and having a really good Twitter list of people on Twitter that talk about AI, those are the two places I pretty much find all of my AI news. They also have a little AI button up here, synthesize articles or extract data with

AI actions. I can click this and get executive summaries and things like that. Actually, haven't played with that feature too much, to be honest. I mostly just kind of scroll my feed every morning to see what news that's come out that I might have missed. I update the Future tools News website, and I pretty much move on with my day after sort of reading the articles that I think are relevant. Next up, we have my mind. So, I used to use Evernote a lot. I would come across articles online, and I would throw them in Evernote. I'd come across YouTube videos, and I would like make a new document in Evernote and save the URL there 'cause Evernote didn't save videos. And I just tried to throw everything into Evernote, but over time, it got to a point where I would throw so much stuff in there, it was really hard to surface old things that I've thrown in Evernote. Then I came across my mind. It's like if you took Evernote and Pinterest and the two had a baby, and then you sprinkled

some AI on top that's my mind. So basically, I've got this little plus button up in my Chrome dashboard. I've also got my mind installed on my phone. So when I come across a YouTube video or a news article or a really cool tool that I want to revisit or anything like that, I click the my mind button either in my Chrome browser or on the actual app on my phone, and it saves it into my mind. And it does it in this like Pinterest-looking fashion where it's a whole bunch of like squares with the thing that you saved. Now, what's cool about this is it's really easy to find what you're looking for. You've got this giant search my mind at the top. So let's say I'm doing a video about stable diffusion. Well, what's all the stuff I've saved recently related to stable diffusion? I can type stable diffusion, and it will pull up any of the stuff I've recently saved related to stable diffusion. Let's say I'm doing a video about clad prompts. I can type in clad and

it will pull up all of the stuff I've saved recently related to clad. It's pretty handy to resurface stuff like that quickly. But other than that, it's also got AI tagging and an AI TLDR. So if I click on this one that I saved earlier, you can see the title of it is California Bill Set to Ban Civet AI, Hugging Face, etc. And then we've got a little TLDR that AI wrote up for us. It says, Proposed California Bill Seeks to Ban AI Image Generation Systems That Fail to Implement Extensive Watermarking Requirements. Critics Argue the Bill Is Technically Infeasible and Represents an Overreach, etc. And then also, it used AI to add these mind tags here. So it added the tag webpage, open source, policy, technology, image generation, California, AI regulation, etc. And if I want to dig into more legislation, let's say I'm going to make a video about this California bill, and maybe I want to mention other legislation that's happening, I can click on this legislation tag, and it will show

me other things that I've saved related to legislation, or let's say I'm making a video about image generators. I can click on this image generation tag, and it will show me other recent stuff that I tagged with image generation. So, it's just a really great tool to resurface stuff that you've come across online again. What it can be? Tweets, YouTube videos, blog posts, GitHub repositories. I save a lot of those here. It can be products. A lot of times, I'll see an ad on Instagram, and I'm like, That actually looks like an interesting product. I'm not going to buy it right now, but I'm going to revisit it. I'll save it to my mind so it pops up to the top of my feed later. Super helpful for people that just want to digitally hoard information but don't want to spend the time to organize it because it does all the organization for me. All right, next up, let's talk about some of the AI image generation that I do. If you look at my YouTube channel

you'll notice that pretty much all of my popular videos use AI-generated images of my face or just, you know, AI-generated images in general. It's sort of become one of my YouTube thumbnail calling cards, I guess. And at the moment, my favorite AI image generation tool is actually still Leonardo. Now, I did a whole video a couple of weeks ago where I broke down the comparison of all the different AI image generation tools. And when I put them all side by side, I kind of mentioned that aesthetically, for me, I just really like the color palettes and the contrast and the quality that I tend to get out of Leonardo the most. Now, it may not be the absolute most realistic that's available on the market right now, but it really fits the style that I'm going for when I'm trying to create my thumbnails and things like that. Full disclosure, I am an adviser for Leonardo, but I do use all of the image generation platforms. And at the moment, the Leonardo Phoenix model is my current favorite model to use. I still

use flux. Sometimes I still use Mid Journey. Sometimes I still use Dolly 3. Sometimes, but the Phoenix models are my main go-to. In fact, if I look at my image creation tab here, you can see a lot of the thumbnails that I've released lately. You'll see the initial concept here. They don't have my face in them yet, but they have the concept before I swapped my face out. And you might even recognize some of these as some of the thumbnails that I've used in the past, you know, with just my face swapped in instead of some of these initial images. Specifically when I'm using Leonardo and I go to image creation over here, I always make sure it's preset to Leonardo Phoenix. This is the model that I think is just like perfect. They have some really cool presets over here as well. My favorite presets to use, I really like the Moody preset. It creates some really cool contrasts. I also really like the ray traced one. It gives these images like this sort of computer-generated look. I also really like to use the Cinematic.

preset a lot, so those are kind of the presets that I'll go between. A lot of times, you'll see me use a prompt like, you know, a man standing in a garden holding a microphone with strawberries behind them. And then I'll go and generate it with Moody. You can see right here, these ones are the Moody preset. And then down here, you can see I'll generate the same prompt but with Ray traced. And then I'll try it with Creative and then 3D render. And so, a lot of times, I'll just use the exact same prompt, and I'll just flip through different presets to get a whole bunch of different variations. And then I'll pick a couple of those to split test my thumbnails on YouTube. I'll give an example here. Let's do a white man on stage with a short brown beard pulling open a curtain to reveal a treasure chest. I always put some of this extra stuff in there so that the face comes out as close to mine as possible if I'm going to do a face.

swap on it. I've noticed if I just put a and well, I might get a skin color that isn't, you know, matching my skin color. So, let's go ahead and generate this. I have it preset at Moody here, and here's a pretty decent image that I can use. It's a little grainy, it's not perfect, but I'll show you another tool that I used to fix some of that in a minute. So, let's go ahead and download this image. And the reason I wanted to download this image is because I wanted to talk about the next tool that I still really love to use. Now, if you go all the way back like 2 and a half years ago on this YouTube channel, some of the earliest AI videos that I was making was around using stable diffusion and automatic 1111. To this day, I still use this tool constantly. Now, I mostly use this tool just to do a face swap. I created a fine-tuned stable diffusion model that actually has my face trained into it, so I can generate images with my face.

that's a little bit more complex of a topic that I'm going to get into in this video, but I did make a full tutorial breakdown of actually how I trained my face into the AI to do that. But essentially, the way I do that is I use automatic installed locally on my computer. This isn't actually a website. I downloaded it, installed it on my computer. I do have tutorials on how to do that as well. And I go to the image to image tab here, and under these secondary tabs here, I'll select in paint. And let's just pull over that image that we just generated inside of Leonardo here. What I like to do when I bring one of these images in is I like to see what stable diffusion thinks this image is first. So I click interrogate clip, and I let stable diffusion actually decide what this clip looks like. So it came up with a standing behind a chest of gold in front of a curtain stage with a spotlight on it and a bunch of extra stuff here. And then

also have a bunch of negative prompts that I like to put in too: ugly, disfigured, cut off, cropped, etc., etc. And then what I'll do is I'll use this little masking feature here and then just mask the face. So let's just mask off the face right here. I'm going to have it replace it with my face. So my little code to do that is actually Dominic Cooper person. Weird, I know. I explain why it's like that in the tutorial about training your face. Then under resize mode, I like to use just resize lat and upscale. And then under in paint area, I like to select only mask because we only want it to change out the area that I masked. Sampling steps, I like to set it at 40, but this is something that you can experiment with. Batch size, I always like to have it generate four variations for me. Den noising strength, I usually bring down to about 7, but it's something I experiment with. CFG scale, I usually bring it up to about 7.5 or 8. CFG scale is kind of how

closely. It's going to follow your prompt instructions up here. The noising strength is essentially how close to the original image it's going to try to keep it. So, you're constantly trying to balance those two to get the ideal image. So, you can just kind of play with those until you get what you're looking for. These are the settings that typically work the best for me. And then if I generate an image here, you can see it made some variations where it is actually me in the image for one of the reasons. It made me look like that in one of them. And this one, it made me bald for some reason. But this one's not too bad. And that one's not too bad either. Usually, I'll do a few rolls until I get it right. So, now we have this variation, this variation, this variation, and this variation. I think this one looks pretty decent. So, I'll go ahead and save this image. And this is the beginning of how I would make a thumbnail. But like I mentioned, it's a little Pixy, a little

grainy. It's not great quality. So, the next tool that I use constantly is this tool, Topaz Labs, their Topaz Photo AI. This is an upscaler that denoises, sharpens, upscales, recovers faces, removes objects, does all sorts of cool stuff. It is a paid product; you pay $200 one time, and then you pretty much have it for life. You don't have to pay any monthly fees or anything like that, but it's pretty cool. I will simply grab the image that I just created with my face on it and drag it in here. You can see here's the original, here's the new one. It zooms in so we can kind of see how it cleans it up a little bit. And if you look at the face, here's what the face looked like in the original when it zoomed in, very pixely. Here's what the upscaled version looks like, a little bit better. It kind of screwed with the fingers a little bit, but other than that.

it pretty much improved the quality pretty dramatically, especially when you start zooming in and looking closely. I'll go ahead and save this image. And there's another upscaler I do like to use from time to time called ca. a. This one does a pretty decent job, but it often makes my face look like a completely different person. If I go to ca. aome and click on enhancer, this one actually will hallucinate a little extra into the image, so it will actually change the look of the image a little bit. So if I pull in my original sort of pixelated image here and upload it, and then come over to the right and click enhance, we can see here's what the original image looked like, and when I pull this over, here's what the enhanced image looks like. You can see it just makes me look like a different person most of the time, which is why I often don't use this one 'cause I kind of want my face to still look like me, but it does an okay job. You can see some of the

other images I've previously done in here. Here's what the original looked like. Here's where the upgraded one looks like. You can see it put some more gaps in my teeth. Here's another one that you might have seen the thumbnail of. There's the original. There's the enhanced version. I actually thought the enhanced version was good enough, so I used it. And I have all sorts of examples of where I've used Kaaya to sort of slightly improve the image. Here's another one. This was the original. Here's the upgraded one. I actually thought the upgraded one was good enough, so I used this upgraded version as well. But it always makes my face look slightly different than reality. Now, another tool that I want to talk about that has legitimately changed a lot of my process with YouTube is I no longer have to go and find stock music to use behind my videos. I can just generate music with Sunno. And another thing I like to do as well is if I'm doing some sort of montage in my video.

and maybe I'm not speaking throughout that Montage. I like to make a song that explains what's happening during that Montage because I think it's just sort of a funny touch to have a song going that's like a poppy song that you want to bob your head to, but the song is also explaining what you're seeing in the Montage. I always thought that was pretty fun. Like, so for example, I was making a video with Gen 3, and I was waiting for Gen 3 to process it, and I'm like, I'm going to put some music over this while it's actually processing it. And so it made this song [Applause] Here No Light the second fight so much. You get the idea. I was able to put this heavy metal song over the Montage of me waiting for Gen 3 to process it. It just adds this little fun, more entertaining touch to some of the videos. Also, sunno is just fun. I've had days where like my wife and kids are

hanging out in my office with me, and we're just throwing ideas at so for silly songs. You can see up here, I have 10 pages of songs that I've generated in here just because I'll get in here and play around with it. Next up is otter, and I've been using otter more and more when I go to conferences. Otter itself actually has an iPhone app where you can open up the outer app, press record, set it in front of you, and it will record the presentation that's going on. It will record all the audio from the presentation, and then when it's done, you can just go and read a summary. So you don't actually have to sit there while you're in one of these conferences or presentations anymore. You can just open up otter, set it next to you, let it record the presentation, and when you're done, get bullet points of exactly what was talked about in that presentation, right? So I was at this Cisco live press session here called AI humanity and the future of work, and you can see here it

basically wrote me a summary of the entire presentation and gave me an outline of all of the various things that were talked about in the presentation. So, I was literally able to kind of check out. I was just hanging out in there. I was watching. I was paying attention, but I didn't have to sit and take notes because I had my otter assistant taking notes for me. Now, I've actually improved this process a little bit. I actually have an external recorder now that I use, and I'll record with this external recorder. That way, I can still be doing stuff on my phone while the presentation's going, and then otter has this import feature where you can go and import the audio recording from your recorder, and it will do the same thing. It will summarize it and give you bullets and all that kind of stuff. So, now I can sit there and scroll Twitter or whatever while I'm watching a presentation at one of these conferences while my audio recorder records the whole thing. At the end of the day, I take the MP3 off my audio recorder, import it.

into OT, and I have a summary of notes from that presentation. Pretty game-changing. And finally, this is a newer one that I started using. It's called Spotter Labs. This one's probably only going to be relevant if you have a YouTube channel yourself, but it's pretty cool. It's an AI tool that helps you spot trends, helps you with ideas, helps you with titles for your videos, helps you with thumbnails, all sorts of really, really cool stuff. So if I go to like Trend Spotter, it'll actually look at the videos that your audience also watched so that you can try to find trends of what people in your niche might already be watching. Right? Here's Peter Diamandis. Did an interview with Ahmad MTO, and it's got 34.6 times the normal views of one of Peter Diamand's normal videos. Here's one from a channel called Cyber Jungle, which I've not heard of that channel before, but we can see they did one about Flux Plus Luras, and it's 13.1 times

bigger than the majority of the videos they post on their channel. It's got 107,000 views on it. Here's one from Matthew Burman AI uncovered. I can look through this to try to find ideas for videos to kind of make my own version of. I really, really like the thumbnail explo. This one to me is really cool. So it says describe a video concept or thumbnail. So let's say I want to make a video about pulling back the curtain on my favorite AI tools. Then click generate, and it's actually going to come up with some thumbnail concepts. Now you're probably not going to want to use the exact image that it creates, but it gives you a concept that then you can go generate with one of the other AI image generators. And this actually looks at your YouTube thumbnails already to see what kind of thumbnails work, and it will give you ideas for similar thumbnails like it just did here. So it gave me four thumbnail ideas, and notice how all of them are a dude with a dark beard. Well, it knows what I look like, right? I can

click refresh all thumbnails. It's going to generate a handful more. I actually really like this concept down here. I'm probably going to roll with something pretty similar actually for my YouTube thumbnail. But now that YouTube allows you to split test three, I can keep on generating more and more and more until I have three good ideas. Make all three of those thumbnails, put them on YouTube. So those are the tools that have legitimately changed my life that use AI. These are the tools that I'm using every single day now. So when that question comes up of do you actually use AI yourself or did you just talk about it? Yeah, I mean my business is really, really run on AI. I use a lot of AI. It helps me come up with ideas for videos. It helps me come up with scripts. It helps me come up with thumbnails. It helps me come up with the idea for the thumbnails. It helps me transcribe presentations that I'm sitting in. It helps me organize my links that I put in the description of my videos. I am trying to automate as

much of what I do with AI as possible. The only thing I don't want to automate is sort of my voice, my opinion, the things that I find interesting to share with you. That's never going to be automated. But when it comes to helping me outline and find analogies, when it comes to helping me with ideas for thumbnails, helping with ideas for titles, all of that, I use AI to help me sort of hone in on what's going to work the best to get the most amount of views and the most amount of attention on my channel. There you have it. That is my AI marketing stack that I am constantly using. If you like stuff like this, make sure you like this video and subscribe to this channel. I make a lot of videos like this. I also keep you up to date with the latest AI news. I make AI tutorials, all sorts of cool stuff in the AI and emerging tech world. So subscribe to this channel if you like that stuff. And also, if you're really into AI tools, check out futur tools. This is a site

that curates all of the coolest AI tools. It's updated every day with new tools that maybe you haven't come across yet. There's also a free newsletter, and the news page is up to date every day. It's a cool site. I think you'll really, really dig it if you like AI tools. It's totally free future tools. And that's what I got for you today. Hopefully, you found this video helpful. Hopefully, there's some ideas you can implement into your own workflows. I really appreciate you hanging out with me, and I will see you in the next one. Thanks so much. Bye-bye.

AI Tools You’ll Use Everyday (And How To Use Them)

AI, No-Code, Tech, Futurism - I'm a tech nerd and talk about tech nerd stuff

Matt Wolfe

저는 다양한 AI 도구를 테스트하고 사용하지만, Claude는 매일 꾸준히 사용하는 기본 도구입니다.

기사 정리, YouTube 제목 생성, 광고 스크립트 작성과 같은 작업에는 Claude의 프로젝트 기능을 활용합니다.

다른 프로젝트에는 코딩 이니셔티브, YouTube 성장 컨설팅, 이미지 프롬프트 최적화, 빠른 인사이트를 위한 기사 요약 등이 있습니다.

콘텐츠 제작을 위한 AI 도구 사용

이 발표자는 짧은 동영상 스크립트와 튜토리얼을 작성할 때 Claude를 사용하며 ChatGPT보다 더 효과적이라고 생각합니다.

Claude의 가장 큰 단점은 인터넷 검색이 불가능하다는 것인데, 이 화자는 최신 정보를 얻기 위해 Perplexity를 사용하며 Chrome의 기본 검색 엔진으로 설정해 놓았습니다.

이 화자는 Perplexity를 사용하여 임의의 질문에 답하고 정보를 수집하는 데 자주 사용하므로 일상 생활에서 필수적인 도구가 되었습니다.

AI 뉴스를 최신 상태로 유지하세요: 도구 및 전략

Feedle은 간단한 RSS 피드 리더로, 최근 주제 기반 웹 알림을 생성할 수 있는 AI 기능인 Leo를 추가했습니다.

AI 피드는 사용자가 보고 싶은 내용을 더 많이 또는 덜 표시하도록 학습시킬 수 있어, 개인화된 뉴스 소스가 될 수 있습니다.

My Mind는 AI를 사용해 저장된 기사, 동영상, 도구를 정리하고 쉽게 검색할 수 있도록 Evernote와 Pinterest의 요소를 결합한 서비스입니다.

저장된 콘텐츠 정리 및 이미지 생성에 AI 활용하기

이 도구는 AI 태그와 TLDR 요약을 통해 저장된 콘텐츠를 빠르게 다시 표시할 수 있도록 도와줍니다.

법률 및 이미지 생성과 같은 관련 주제 사이를 쉽게 탐색할 수 있는 태그를 제공합니다.

사용자들이 가장 선호하는 AI 이미지 생성 도구는 레오나르도이며, 피닉스 모델은 썸네일 생성에 가장 많이 사용됩니다.

이미지 생성 및 얼굴 스와핑을 위한 AI 도구 사용

이 스피커는 다양한 사전 설정을 사용하여 다양한 이미지 변형을 생성하여 YouTube 썸네일을 만듭니다.

또한 자신의 얼굴로 모델을 훈련시킨 후 안정적인 확산과 자동 1111을 사용하여 얼굴을 교체합니다.

이 과정에는 생성된 이미지를 다운로드하고, 안정적 확산을 사용하여 얼굴을 조사하고 마스킹하며, 최적의 결과를 위해 다양한 설정을 실험하는 것이 포함됩니다.

썸네일 만들기 및 이미지 개선하기

이미지 생성 시 노이즈 강도는 원본 이미지와 변형된 이미지 사이의 균형을 맞추는 데 도움이 됩니다.

토파즈 랩스의 사진 AI는 이미지를 업스케일링, 노이즈 제거, 선명하게 처리하여 품질을 획기적으로 개선하는 일회성 유료 도구입니다.

a의 인핸서도 이미지를 업스케일링하지만 피사체의 외관을 변경하는 경우가 많기 때문에 일관된 결과를 얻기에는 신뢰성이 떨어짐니다.

동영상 제작 및 노트 필기를 위한 크리에이티브 도구

이 스피커는 음악을 사용하여 비디오 몽타주에 재미를 더하고, 처리를 기다리는 동안 종종 우스꽝스러운 노래를 만들기도 합니다.

Otter의 앱은 컨퍼런스에서 프레젠테이션을 녹음하는 데 사용되어 나중에 요약과 요점을 제공하며, 발표자는 외부 레코더를 사용하여 프로세스를 개선했습니다.

Spotter Labs는 YouTube 크리에이터를 위한 AI 도구를 제공하여 트렌드를 파악하고, 동영상 아이디어를 생성하고, 썸네일을 만들 수 있도록 도와줍니다.

AI를 사용하여 YouTube 콘텐츠 제작 향상

연사는 AI 도구를 사용하여 동영상 아이디어와 썸네일 컨셉을 생성한 경험을 공유하며 워크플로에서 자동화의 중요성을 강조합니다.

비디오 스크립트 작성부터 링크 구성에 이르기까지 비즈니스의 다양한 측면에 AI가 미치는 영향을 강조합니다.

또한 AI 도구를 큐레이팅하고 업데이트하기 위한 웹사이트인 퓨처 툴을 홍보하며 무료 뉴스레터와 매일 업데이트를 제공합니다.