// transcript — 532 segments
0:00 <Untitled Chapter 1>
0:01 We've got Mind Sweeper. Does it actually work? Oh my gosh. Yeah, Mind Sweeper
0:06 actually works inside of this Windows XP simulator. As somebody who looks at and
0:11 tries out AI tools every single day, I am constantly on the hunt for the
0:15 absolute best free tools that people can use right now. One, because well,
0:20 they're awesome. And two, because who knows how long some of these are going
0:22 to be free for. Some of these companies have a tendency to let you try them for
0:26 free and then once you're hooked on using them, well, then they start to
0:29 charge you. That's pretty clever, Dad. So, I want to show them off now so that
0:33 you have the opportunity to try them out while they're still free. And let's
0:34 Google AI Studio
0:36 start with the free AI tool that I probably use more than any of the rest,
0:40 and that's Google's AI Studio. You can find it over at a studio.google.com.
0:44 And well, I'm not going to go too deep into this one cuz I literally made an
0:49 entire breakdown video of all of the cool things that you could do for free
0:53 with Google AI Studio. So, definitely check that video out if you haven't
0:56 already. In that video, I used their free AI code generation tool to create
1:01 this emoji fusion master game where you can combine elements and it turns them
1:07 into new elements like celestial fluff. I created this alien dodger game which
1:11 works a lot like Frogger, but you're an alien. And I even completely cloned one
1:16 of my favorite apps, the Feedley app, all using Google's AI Studio coding
1:20 tool. Inside of the free AI studio, you can also do text to speech. You can
1:25 generate images. You can create music in real time. And you can play around with
1:30 pretty much any one of Google's AI models that are available out there.
1:33 Another really cool feature that you can do, again, totally free inside of AI
1:38 Studio, is that you can share your screen or share your webcam while having
1:42 a conversation with the AI, and it will actually walk you through how to do
1:46 things. So, for example, I could look into my YouTube analytics here and get
1:51 advice from the AI. Based on my YouTube analytics, what types of videos should I
1:56 focus on if I really want to grow the amount of subscribers my channel has?
2:01 Based on your analytics, the video that brought the most subscribers is the one
2:05 about 50 plus insane new ways to use the new chat GPT. And also, the AI news
2:09 updates seem to attract quite a few subscribers as well. Focusing on similar
2:14 topics like AI news and practical uses of specific AI tools could help you grow
2:18 your subscriber base. It's super cool. And I could come back
2:22 and see exactly what the AI saw. I could listen back to what I asked it and I can
2:26 NotebookLM
2:26 listen back to the response that it gave me. And since we're talking about
2:30 Google, I need to show off another Google tool that's totally free to use
2:34 that I absolutely love and that is Notebook LM. This is a tool where you
2:38 can upload a bunch of documents or website URLs or even plug-in videos and
2:43 then ask it questions about those videos. It'll make mind maps about those
2:46 videos. But the feature that I like the most is the ability to actually create a
2:51 podcast around the information that you feed it. So recently I got a new camera
2:56 called the Fujifilm X1006 and I wanted to learn how to use it. So
3:00 I went and found a handful of resources all about it. plug those resources into
3:05 notebook LM and we can see that it created a mindm map for us around it
3:09 right here and I can dive deeper and deeper into this camera's features and
3:15 you know technical specs and things like that by digging deeper and deeper into
3:19 this mind map that it generated for me. But it also generated this podcast that
3:23 essentially explains more about the camera so I can listen to a 20inut
3:27 podcast and get a deeper understanding of this new camera that I just
3:30 purchased. Have you ever looked at a photograph, maybe one taken with, you
3:34 know, an incredible camera like the Fujifilm X1006? And the truth is, it's
3:39 probably not the camera. It's more about really knowing how to unlock all that
3:42 potential packed inside it. And I use this Notebook LM tool all of
3:46 the time. It's so amazing when I'm trying to dive deeper on a specific
3:50 topic to just throw a ton of resources in here. You simply click this add
3:54 button. You can throw in PDFs, text files, markdown files, MP3s, PGs, JPEGs,
4:00 YouTube links, Google Drive files, or just copy and paste text. And once it's
4:04 in there, you get the mind map, you get the audio podcast, and you can have chat
4:08 conversations with all that information and ask things like what essential
4:12 settings and techniques optimize the X1006 for diverse shooting scenarios.
4:17 And we get a response with a lot of details and the exact source. So, if I
4:21 click into one of these links, it'll actually pull the exact document it
4:25 found it from and highlight the section that it actually got that information
4:28 in. This is super powerful and still blows my mind that they just let anyone
4:31 Veo 3: Ask Perplexity Bot
4:32 use this for free. And we can't talk about Google without talking about
4:37 Google's V3. V3 is their new video generation model, which is really great.
4:41 It's really realistic. It can generate audio. All of the memes, all of the
4:45 crazy viral AI videos you're seeing lately were probably generated with V3.
4:50 Now, VO3 not free to use. Google actually charges quite a bit to use it.
4:55 However, there is a trick that's working right now. Can't guarantee this will
4:58 work forever, but right now you can actually generate videos for free with
5:05 VO3 using Perplexity's Xbot. So, if you go over to X and you submit a prompt on
5:11 X where you tag ask Perplexity, you can actually give it a prompt and it will
5:14 make that video using V3. Now, it's using V3's fast model, so it's not their
5:19 highest quality model, but it's still pretty dang good. So, you can see here I
5:22 went, "Ask perplexity, make me a video of a talking cat that's sitting next to
5:26 a taco and saying, subscribe to Matt Wolf." And here was the result.
5:31 Subscribe to Matt Wolfie. Now, it was supposed to be the cat
5:35 saying that, but it's AI. Maybe a couple more attempts and it would have gotten
5:38 it right. But that is a way that you can go play with V3 right now for free. I
5:42 Kling
5:42 mean, depending on when you're seeing this video. And since we're talking
5:46 about AI video, let's talk about Cling. Cling is probably one of the better AI
5:51 video generators available on the market right now. And they do have a free tier
5:55 where you can generate videos for free. Now, they take a little bit longer to
5:59 generate. They do prioritize paying users, but you can generate videos for
6:03 free with this tool. If I go to clingai.com and then come up and click
6:07 on create, it will jump me into the Clling platform where I can generate
6:11 videos. Now, if we look at their pricing here, you can see that for $0, you get
6:16 credits by logging in. So, the more often you log in, the more credits you
6:20 get, and those credits can be used to generate videos. I logged into a free
6:24 account. You can see this is not a paid plan. It's telling me to upgrade. And it
6:29 gave me 166 credits to play with. I could give it an image like this one
6:33 with a woman and a cat. And give it a prompt like, "A woman gets closer to a
6:36 cat and the cat gently sniffs her. a gentle breeze brushes her hair, the
6:39 cat's fur and the flowers in the background as clouds drift away and the
6:45 camera zooms in. And you do have options for other camera effects as well. It
6:49 will also add sound effects to the videos, very similar to VO3. And we can
6:53 see it's going to use 20 credits. Again, I've got 166 to play with. Let's go
6:57 ahead and generate. And we get a video You can also do text to video. So, if I
7:07 want to use my a monkey on roller skates So, again, that's Clingai. You can find
7:18 it at clingai.com. The video generations aren't the fastest if you're on the free
7:22 Leonardo
7:22 plan, but I mean, it's pretty good for free. Now, let's talk about image
7:26 generation. My favorite image generation platform by far is Leonardo AI. Now, I'm
7:32 a little biased. I do have some equity in the company, but I have equity cuz it
7:36 was a company I loved. In my opinion, they have the best platform and user
7:41 interface for generating AI images, and their new Lucid Realism model is
7:46 probably my current favorite AI image generation model. Now, I do have an
7:49 upgraded account on Leonardo, but I'm going to log out and log in with a free
7:53 account so I can show you what you actually can do for free in Leonardo.
7:57 So, this is a brand new Leonardo account here. I'm not a paid user. This is
8:00 completely free. You can see that it's telling me I'm on the free plan and I
8:03 need to upgrade if I want priority generations. But I have access to these
8:09 models, including the GPT1 image generation model here, the Flux context
8:14 model, and my current favorite, Lucid Realism, which is an exclusive model
8:18 just inside of Leonardo. I'm going to have it generate a random prompt for me.
8:22 See what it comes up with. And it's going to cost me 66 credits to generate
8:27 this. You can see I've got 150 for free right now. and that is refreshed every
8:31 day. Also, keep in mind I can use less credits by making the images smaller or
8:36 by switching to different models. Different models have different amounts
8:39 of credits they use, like the Flux Schnell model will only cost me eight
8:43 credits to generate images with it. The Leonardo Phoenix model only 20 credits
8:47 to generate four images. But again, I really like this Lucid Realism model.
8:50 So, let's go ahead and generate the suggested prompt here. And you can see
8:54 the quality of images that it generated here. They look pretty good. I feel like
8:58 it's hard to get a real feel for the style when it's more of this like anime
9:01 style. So, let's try to get something a little more realistic. I'll just do a
9:04 man stands on the sidewalk in front of a busy city skyline. We'll go ahead and
9:08 generate. It's actually going to improve that prompt for me. You can see how it
9:11 just changed the prompt to give it a little bit more detail. And we get some
9:16 solid, pretty realistic looking images. You can even do things like edit the
9:19 images similar to what you can do in OpenAI. So, if I click into one of these
9:24 images and say put a baseball cap on the man and click generate, we can see that
9:29 it generated an image. It's the same image, but now it put a baseball cap on
9:33 the man. Now, I'm pretty much out of credits for the day and can come back
9:37 tomorrow and use more credits. Or next time I can try different models that
9:41 Thiings
9:42 also use less credits. Here's another fun one that I recently came across
9:46 called things. It's actually spelled T H I I N G S and it's a whole bunch of AI
9:53 generated icons for pretty much anything you can imagine. And if it doesn't have
9:57 one that you want, you can actually request that they generate it. Now, you
10:01 can generate your own things, but it costs credits and they don't seem to
10:05 have any sort of free aotment of credits. But you can suggest a thing and
10:09 if it's not in their database already, they'll create that thing for you. And
10:13 chances are whatever you're looking for an image of, it's probably going to have
10:18 it. Like if I search a wolf, there's definitely a handful of options
10:22 available to me. If I search a monkey, we even got monkeys smoking cigarettes.
10:27 We got tacos, various animals, various vehicles like drones that look like the
10:32 DJI drones. I mean, there's probably a good chance you're going to be able to
10:34 find the icon for whatever it is you're trying to do in here, and it's probably
10:38 already been created. Pretty cool resource that I thought would be fun to
10:39 Whisk
10:42 share. Next up, let's talk about Whisk. This is another one out of Google that's
10:46 free to use and it allows you to sort of combine concepts into images and I found
10:50 it to be really fun. Let's go ahead and click enter tool. This one you can find
10:59 Let's go ahead and enter the tool here. And if we open up the left sidebar, we
11:04 have various elements that we can tweak like the subject, the scene, and the
11:08 style. So for the subject, I like to upload an image. I'll upload an image of
11:13 myself here. And then under style, you can either enter a text, upload another
11:17 image, or we can do something random like here's an anime looking image. And
11:22 then for scene, let's just put in space. So we've got me as the subject, sort of
11:27 space is the scene, and then in an anime style. And I can go ahead and click
11:30 generate. And it mixes all those three together to make an image of me in space
11:36 in an anime style. They've also got some fun presets. So, if I click on the
11:40 little hamburger menu here, there's some presets like a sticker, enamel pen,
11:44 plushy, card, chocolate box, capsule, toy, bento box. I've always thought the
11:48 plushy was kind of fun. Give it an image of myself. Let's get rid of the
11:52 background scene. And then let's use style plushy and just use those two and
11:58 click generate. And now we have a plushy me. Now, there are buttons here to
12:01 animate these if you want, but I believe this animate button is only going to pop
12:04 up if you're on one of the paid plans. But the ability to sort of generate
12:09 images by combining images that you can do for free. One of these days, I really
12:12 Glif
12:12 want to get one of these plushies made. I think it'd be really fun. Next up, I
12:16 need to tell you about Glyph. Glyph is this like AI workflow builder where you
12:22 could sort of tie together different AI tools to get a different outcome. And
12:26 then you can share those glyphs so other people can use them. And you know what?
12:30 It's probably best that I just show you. So if I go to glyph.app, that's
12:35 glif.app, app. I can click on explore and we can see some of the other glyphs
12:39 that people have made. So, for example, someone made a YouTube thumbnail maker.
12:44 I can click into here. It asks for a prompt. The example, an Android robot
12:47 smiling and showing a phone screen which says glyph.app. Sure, let's run this
12:52 glyph. It's going to use 4.12 credits. You actually get an aotment of credits
12:56 every single day. I actually currently have a,000 credits and I've never paid
13:00 for them. You just get 10 a day and they seem to stack up. And we can see here,
13:04 this is the workflow for this glyph. We gave it a prompt. It's actually
13:08 enhancing the prompt. It's setting the width, the height, the steps, and then
13:12 it's doing a bunch of comfy UI magic stuff in the background using AI to give
13:18 us our final YouTube thumbnail. Now, not my favorite YouTube thumbnail I ever
13:22 came across, but not bad for just giving it a single prompt. Let's check out some
13:25 of the other glyphs that people have made. Like, here's a glyph called
13:28 Spiderverse. It hallucinate in the style of the Spiderverse movies. So, let's put
13:33 I don't know. Let's just put Batman. Run this glyph. And I guess this is what
13:36 Batman would look like in the Spider-Verse. Probably pretty accurate.
13:40 Looks good to me. But where glyph is really powerful is building your own
13:42 workflow. So, let's go ahead and click on build here. And you can see you get
13:47 like these glyph blocks. So, let's go ahead and start building with glyph
13:50 blocks. We want to start with a text input. This is where somebody could
13:54 enter a prompt. Let's reword this to enter your prompt. And then we want it
14:00 to take that prompt and maybe use a text generator to improve the prompt. We
14:04 could select any of these models here, which is pretty much every available
14:10 model. Let's use GPT40. And we can give it a prompt like improve the following
14:17 image prompt so that it has brilliant colors and beautiful lighting. Also add
14:22 a wolf somewhere in the image prompt. There should be a wolf in the background
14:25 of the image. And then we'll go ahead and give it the input. And then I'll go
14:29 prompt and it knows that this is the prompt to improve. So now you enter an
14:35 image prompt. It'll run through GPT40. Improve the prompt. Add a wolf to the
14:39 prompt. And now we want it to actually generate an image. So we'll use an image
14:44 generator. And then for our image, we're going to pull in our text here. So we'll
14:48 use text one. And then we can select what type of image generator we want to
14:53 use. I'm going to use Flux Pro version 1.1 ultra and let's make it 16 by9. And
14:58 now we can test this glyph. Let's just do Batman again. Right. So my prompt is
15:03 just Batman, but this is going to make it colorful and add a wolf to the image
15:07 with every single image. Let's run this glyph. And just by giving it the prompt
15:12 Batman, we got a beautiful colorful scene with Batman and of course a wolf
15:17 somewhere in the image. And I can give it any prompt I want and it will give it
15:21 those additional details to add the brilliant color and to add the wolf. So
15:26 I can call this glyph give me a wolf. Then I could publish this glyph and
15:29 anybody can go use it now. Now if I go back to my explore page and look for
15:35 new, you can see my give me a wolf glyph is available for people to use. But this
15:39 isn't just for image generation. You can have it generate videos if you know what
15:41 you're doing with Comfy UI. You can build out entire Comfy UI workflows. You
15:46 can have it generate audio. You can have it take an image and take it down to
15:50 text. You can make audio to text. You can extract JSON. You can make HTML. You
15:55 can give it a video input. You can have it generate videos. All sorts of
15:58 workflows. And you can make these workflows as long and as complicated as
16:02 you want. And again, you get a certain amount of free credits every day. And
16:04 ElevenLabs
16:05 you can just play around with this stuff. Next up, let's talk about 11
16:09 Labs. This is probably the best audio generator available, especially for text
16:13 to audio. And you can do a lot with it for free. Now, I do have a paid plan,
16:17 but I'm going to log into a free plan where I'm not paid on this account and
16:21 show you what we can do for free. Once I'm logged in, I could use their text to
16:25 speech generator where they have tons of voices that I can use. Like, Alice,
16:29 just trust yourself. Arya, there is no greater harm. Bill, as we are liberating, Brian, the
16:36 thing always happens that you really believe in. Let's go ahead and use Arya and generate
16:42 some speech. Subscribe to Mattwolf on YouTube and check out futuretools.io.
16:48 Not bad. Again, this is on a free plan. You can also use the voice changer where
16:52 you upload an audio file and have it change your voice to a selected voice.
16:56 And you can generate sound effects. So, if I need a sound effect of a car
17:01 crashing through a glass window, I can generate that. And it gives me four
17:10 And there's a ton of other features in 11 Labs that you can use, but some of
17:14 them once you start to use too much credits, you will have to pay, but you
17:17 Suno
17:18 can still do quite a bit for free. Since we're talking about audio, let's talk
17:23 about Sunno. So is a tool that creates AI generated music, and in my opinion,
17:27 it creates some of the best AI generated music. Now does have both free and paid
17:32 plans. If we scroll down here, we can see that on the free plan, you get
17:37 access to version 3.5. So, you don't get the absolute latest state-of-the-art
17:42 music generation model, but 3.5 is still pretty dang good. So, when I'm logged
17:46 into Suno here, I can literally just describe the type of song I want. For
17:51 example, a pop punk anthem about subscribing to Matt Wolf on YouTube. I'm
17:54 going to go ahead and set this to version 3.5 just so you can get an
17:58 example of the type of music it will generate on the free plan. And let's go
18:02 ahead and create. And after just a few seconds, we've got two options to choose
18:07 from. Subscribe to please the guy you need with these sceed at lightning
18:18 And here's the other song. Lightning speed. [Music] All right. So far, we've talked about
18:28 Perplexity
18:34 various creative pursuits that you can use AI for. Well, now let's get into
18:39 productivity because there are a ton of free AI tools that will make your life
18:44 so much easier. One of them for me is Perplexity. I love Perplexity so much,
18:49 in fact, that I did an entire video about all of the cool ways that you can
18:53 use Perplexity. So, if you haven't checked out that video, make sure you
18:56 check that one out. Like most of the tools on this list, this does have a
19:00 free plan and a paid plan. I personally do subscribe to the paid plan and love
19:04 it, but I'm logged in on a free account here. This is not a paid plan to show
19:09 you that you can still get great results on the free plan. For me, Perplexity has
19:14 pretty much entirely replaced Google search. Anything that I would have
19:17 searched on Google in the past, I just search Perplexity now instead. And I
19:21 literally just find myself asking it random questions all day long. For
19:25 example, I was flipping through the channels one day and there was a Texas
19:29 A&M football game on and one of my kids said, "What does Texas&M mean?" And I
19:33 went, "I don't know. Let me ask Perplexity." And with a quick little
19:37 prompt, well, it stands for agriculture and mechanical. I really like Perplexity
19:41 because it essentially goes and searches the web for you. And then once it
19:46 searched the web, it goes and summarizes the results and then even cites its
19:50 sources. So, if you're kind of uncertain with the information that it gave back,
19:54 you could always go and check the source of where it got that information. This
19:58 was obviously not the most complex question to ask perplexity, but it works
20:00 MindStudio
20:02 well for demo purposes. Next up, I want to show you Mind Studio, another tool
20:07 that I've used so much that I've literally made multiple other videos
20:11 about how you can use Mine Studio. But if you're not familiar, there's really
20:16 two main ways to use Mine Studio. You can build these visual workflows in
20:20 here. Kind of similar to what we did with Glyph, but I feel like it's way
20:25 more powerful and has way more various integrations that you can tie into it. I
20:29 could do things like create a user input here. This would be where you would give
20:33 it like a prompt. And then once you give it a prompt, there are so many different
20:36 things you can do from here. You can have it generate an image, generate a
20:41 video, a chart, create a podcast, scrape a URL. Now, some of these features I
20:44 believe are only available on the paid plan, but a lot of this you can still
20:50 build out for free. It can fetch info from YouTube videos and YouTube comments
20:55 and scrape Instagram mentions and search blue sky posts and update Google Sheets.
20:59 It's similar to what you might get out of a make.com or an N8N. And honestly,
21:04 really, really powerful when you build some workflows. But instead of showing
21:08 you building a workflow, let me just quickly show you inside of an existing
21:11 workflow. You can see that the ones that say pro agent, those will require a paid
21:15 plan to actually use. So, let's pick one that's not. So, we've got find product
21:19 alternative sideby-side product and service comparisons. Let's click into
21:24 this one here, make a copy, and we can see what this workflow looks like. This
21:27 is one of the workflows that's available for free. So, you have a user input.
21:31 It's going to ask you for an intro, a name, and a URL. It's then going to
21:36 scrape that URL. It's going to analyze the product using Llama 3.17B.
21:43 Then it's going to find and analyze competitors using Perplexity. Then it's
21:48 going to generate a side panel report. Then it's going to display the report.
21:52 And here's what that looks like in practice. If I click publish and then
21:56 open the AI agent, it's going to ask for a product name and the product homepage
22:00 URL. So let's go ahead and put Future Tools and the homepage URL.
22:05 futuretools.io. We'll click next. And then it goes and runs those processes in the background.
22:11 And after roughly a minute or so, it came back with this report that it
22:15 generated for us. Future Tools, a platform that collects and organizes AI
22:19 tools to help users become superhum. It lists off the key features of the
22:22 website. And then it shows some competitor websites like top tools.ai,
22:28 similar web, favored, and there's an AI for that. also sharing strengths and
22:33 weaknesses of each of these competitors. But Mine Studio also has a Chrome
22:36 extension that's pretty cool. You can see it adds this little icon on the
22:41 right sidebar and it allows me to access various agents directly inside of my
22:46 sidebar. So to show what this extension can do, I opened up a TechCrunch article
22:50 here all about Sam Alman and the New York Times. I can use one of these quick
22:55 bots here to either give a TLDDR or an executive summary. Let's go ahead and
22:59 run the TLDDR bot. And we can see that it gave me a quick summary of what this
23:04 article is all about in my little TLDDR summary here. Again, this is just
23:08 scratching the surface, but there's all sorts of little bots that you can run on
23:12 various web pages to get extra information or to automate processes
23:17 URL to Any
23:17 based on the content of the page and things like that. It's actually really
23:20 really cool. Here's another one. Super super simple concept. It's called URL to
23:24 any. You give it a URL. It'll give you back either a markdown file or an HTML
23:31 file or a PDF or an image or a JSON or a QR code or whatever you want from that
23:35 initial URL. So, let's go ahead and go back to this TechCrunch article we were
23:39 just looking at. I will copy this article URL and let's say I want a PDF
23:44 version of this URL. I'll do URL to PDF. Plug in my URL and click convert. Within
23:50 about two seconds, I now have a PDF version of this article right here that
23:55 I can download and use however I want. Just something to bookmark that you'll
23:57 11.ai
23:59 probably find handy from time to time. All right, now let's get into even more
24:03 of the like AI agent kind of stuff. This is a brand new one that just recently
24:08 rolled out called 11.ai. You can find it over at 11.ai. It's actually created by 11 Labs, but
24:14 it's designed to create a personal voice assistant similar to like Siri or Alexa
24:20 that ties into a lot of the products and online services you use. And if we click
24:24 this little learn more button here, is 11 AI free to use? Yes, all calls
24:28 through 11AI are free during the alpha period. So, as of right now, this is
24:33 free to use, but again, this may not be free forever. So, we can go ahead and
24:37 name our assistant. We'll just leave 11 as the name. We can pick a voice that we
24:43 want to use. Let's do Bert Reynolds for the good old American life.
24:53 Now we can add some integrations here like our Google calendar, hacker news,
24:58 linear, notion, perplexity, and slack. I went ahead and added my Google calendar.
25:03 So if I click start call, hello there. Feeling productive? Let me
25:07 know how I can help. What do I have on my calendar for next week?
25:11 For the week of July 6th through July 12th, it looks like you have Adobe VIP
25:17 Gen AI community call on Thursday, July 10th from 9:00 a.m. to 10:00 a.m. And
25:23 news roundup also falls on Thursday, July 10th from 1:00 p.m. to 2:00 p.m. Is
25:28 there anything else I can help you with regarding your schedule? Now, I imagine
25:31 they're going to roll out a lot more integrations over time, so it can
25:35 connect to even more services, but if you know what you're doing, you can
25:38 actually connect custom MCP servers and have it connect to other services using
25:43 their MCPs if they've got them. Something brand new that just rolled
25:47 out, but pretty fun to play with and over time will probably be more and more
25:50 useful, especially once you start using it on things like your phone and maybe
25:54 Emergent
25:54 if they add it into more devices that you might have around your house, like
25:57 your Alexa. All right, another AI agent that I've been hearing a lot about
26:01 lately is called Emergent. This is an agent that will actually go and write
26:05 code for you. And unlike some of the other tools out there like Cursor or
26:10 Windsurf, which is an IDE where you sort of create the code and then sort of go
26:13 back and forth with it, this is more of an agent where you tell it what you
26:18 want. It will create a task list of things to accomplish and then check each
26:22 task off and then even go back and doublech checkck its work. So, let's go
26:27 ahead and test clone Spotify. We can see that it is setting up an environment and
26:31 it's going to take about 1 to 3 minutes. We can now actually see it running. It
26:35 said I'm starting to run the task to create a Spotify clone. Running a web
26:40 search with query Spotify web player. We can see the description that it came up
26:43 with. Now, let me get some images for the Spotify clone. It's now finding
26:48 images or creating them. I'm not sure. Now, I went away, watched a YouTube
26:52 video, came back, and it says, "Your Spotify clone is done." So, if I scroll
26:58 up, you can see it started this at 9:56 p.m. If I scroll down, it was done at
27:03 10:02. So, it took about 6 minutes to run this whole process. And we can see
27:07 all of the steps that it went through to create this for me. And I can click on
27:12 preview here. And here's my Spotify clone. Let's even open it in a new tab
27:15 to see what it looks like in full screen. It's got a search. It's got a
27:20 your library, but it didn't go far enough for me to actually click into any
27:24 of these yet. So, there's no obviously real songs on here. And building this
27:30 app here used 1.294 of my credits. And you can see up here, I started with 10
27:34 credits total. That's how much you get on the free plan. And I still have 8.71
27:40 to continue to further build out this app with Emergent, a pretty cool AI
27:44 Claude
27:45 coding agent that you can use totally for free right now. Anthropics Claude
27:49 also rolled out a feature called Artifacts which allows you to write
27:54 apps, use the apps directly inside of Claude and then even share access to
27:58 those apps with anybody. And there's even like a little marketplace kind of
28:01 thing here where you can see some of the apps that have been developed. So let's
28:06 try this join dots game here that somebody made. I could click on view
28:10 full chat and I can see the entire conversation that was used to create
28:14 this game. Over here, we can see the original prompt with all the details and
28:19 all of the steps they went through with Claude to produce this app. And we can
28:23 even test the app directly inside of Claude. So, let's go ahead and do easy
28:27 difficulty. And it looks like it's basically connect that I'm playing with
28:30 a computer. So, I put my pink piece in, they put their yellow piece in, I'll put
28:34 my pink piece here, they analyze the board, and they block me over here. I
28:38 drop another pink, they analyze it, and they do a good job of blocking me.
28:43 Pretty cool little app. And you can use these anthropic artifacts for free right
28:48 now. And you can even create your own. If I click on the artifacts item in the
28:51 left sidebar, I can click on new artifact. Let's do a productivity tool.
28:57 And then it asks some questions about the productivity tool that I'm going to
29:00 build. I'm just going to tell it I want a tool that can track streaks on daily
29:03 habits so I can chain days together. And let's see if it gets what I'm getting
29:07 at. I'll go ahead and answer its questions here with additional context.
29:11 And I am purposely trying to give as little detail as possible so I can see
29:15 how well it will perform. Now it's shifted into coding mode and we can see
29:18 that it created this split screen where it's now writing the code on the right.
29:23 And after less than a minute, we've got our habit tracker. So habit street
29:27 tracker. Let's add run as one of the habits. And it put a bunch of giant
29:32 boxes as the streak. So if I do it today, it fills it with green. And if I
29:36 do it the next day, it fills it with green. Not quite what I was looking for,
29:40 but I mean, not too bad on a single prompt. I gave it a few more details
29:45 that I wanted the boxes to be smaller and horizontal. And now my streak
29:49 tracker looks like this. A little bit closer. Not amazing, but not bad for
29:54 three prompts on a free tool either. Oh, yeah. And I can actually publish this
29:58 now and make it available for other people to use the app that I made. No
30:03 need to deal with hosting or setting it up on some other server somewhere else.
30:06 Minimax Agent
30:06 It could just be hosted right here on cloud servers and I could share it with
30:09 people. Now, let's talk about the new agent in town, which is the Miniaax
30:15 agent from Halu AI. Once again, this is a true agent where it will create a
30:19 plan. It will work through its plan. It will go back and doublech checkck its
30:22 work, even adding additional things to that plan until it realizes that it
30:27 concluded what it set out to do. I'm actually going to go ahead and test this
30:30 agent with something that I've been looking for help with. I'm actually
30:34 trying to buy a new truck and the truck I've been wanting to buy, I've been
30:38 having a hard time finding near me. So, I'm going to see if it could do this
30:42 research and find me this truck. I'm looking to buy an F150 Lariat with a
30:45 crew cab. I gave it details about the type of engine I want, the colors I
30:49 want, necessary feature, and then I said, "Please help me find options
30:52 within 600 miles of me. Sort your findings by closest to San Diego to
30:56 furthest from San Diego since that's where I'll be coming from." So, after
30:59 about 30 minutes of running, yes, this is actually been running for quite a
31:04 while, you could see all of the various things that it did to get me this
31:09 listing here. In fact, I was watching it a little bit. It even had errors in the
31:13 middle, recognized its own errors, and then said, "Let me fix those errors."
31:17 And then continued on with the process. And then when it was all said and done,
31:20 it gave me this giant list of information, including giving me the
31:25 search results in multiple file formats. So, let's go ahead and click on the PDF
31:29 format. And there we go. It found a bunch of trucks for me in my area and
31:35 sorted it by distance from closest to me to furthest away from me. It even pulled
31:39 in the phone numbers of the various dealerships, gave me a list of questions
31:44 to ask the dealerships and much more. I mean, this is super valuable. It did
31:48 take quite a long time. These agents that actually work like real agents are
31:52 designed for you to give it a task and you literally walk away while it goes
31:56 and figures out that task for you and Miniax can do that and you get a certain
31:59 Spawn
32:00 amount of free credits to just go try it. I didn't pay for this. Now we're
32:03 getting into some just really interesting sort of fun tools that are
32:07 out there and this is one that I came across called Spawn over at spawn sp
32:12 awn.co. I don't know if you remember the site Stumble Upon where you just keep on
32:16 clicking a button and it would just give you a random site to try and you just
32:20 keep on stumbling around the internet until you find stuff that was
32:23 interesting. This site kind of reminds me of that. It's a whole bunch of AI
32:27 generated games and I think there's some apps in there but it seems to be really
32:30 focused on games and I can just click next destination and it's going to throw
32:34 a new game in front of me like this start snake and I can start the game and
32:38 we have pretty much what looks like a snake game where he collects a banana.
32:42 All right, that game didn't do it for me. Next Destination throws a random
32:45 game in front of me. This time it looks like it's a AI chess game. Here's one
32:49 that instantly looks like a Minecraft kind of game, but way more messed up.
32:55 And I could literally just click through and find little apps that have been
32:58 made. Here's one where I could increase the trail length and it changes the
33:02 design a little bit. Here's some sort of shooting game that I don't know how to
33:05 make it work. Oh, there we go. But you can also use this site to create games
33:10 as well. So, if I come up to the plus button, I can create a new app by giving
33:14 it a prompt or an image. I'll say make a game that's somehow a mix between Tetris
33:19 and Snake. Apparently, it's opening portals to new worlds. And now it looks
33:23 like it's building several different potential games. Tetris pieces that move
33:28 like a snake will be completed in about 16 minutes. Snake game played on a
33:34 Tetris grid with stacking mechanics that'll be ready in 17 minutes. navigate
33:40 Snake through falling Tetris pieces. Snake that leads Tetris blocks in its
33:45 wake. It's apparently setting all of these up. Just kind of running behind
33:49 the scenes. And I should be able to come back and play these various games in
33:55 well about 15 or 16 minutes. I just fast forwarded for you and our four apps are
34:00 done here. All right, we've got Falling Snake. So, okay. So, we've got a snake
34:08 that has to play the snake game while also dodging falling Tetris pieces.
34:15 So, I'm assuming if I hit a Tetris piece, I die. Yep. And then with this
34:19 game, if I want to share it, I can copy and share the link to it. And they'll
34:23 even sell me a domain name to it if I want. For reference, the Falling Snake
34:27 one was the only one that actually worked. These other three all kind of
34:28 WebSim
34:31 had issues that made them unplayable. Along similar lines, we have WebSim,
34:35 which can be found at websim.com. And this is almost like a YouTube for apps
34:41 where people use WebSim, create little apps, and the site will just randomly
34:45 recommend a bunch of different apps for you to play. There's things like Windows
34:51 Simulators, Grow a Garden Game, Geometry Dash, Face Maker, Throw Things at Timmy,
34:57 a Google logo maker, just the most random stuff. World War II Simulator.
35:00 And just like the last site we were looking at, you can create your own
35:03 apps. And over on the right side here, you can choose your model and prompt it
35:08 to create whatever app you want. I think for me though, my favorite part about
35:12 WebSim is just exploring and getting lost in the various random apps that are
35:16 in here. If we go to the top apps, we could see some of the most popular ones.
35:22 YouTuber Simulator, Find the Fake, Variant Generator, just Multi-Pendulum
35:27 Simulation. You can just click into any one of these and just have some fun.
35:31 Like we can mess with gravity on it and see how it affects the thing. Change the
35:36 speed. Change the number of pendulums. This is wild. Here's one called everyone
35:41 versus everyone. So, choose my three fighters. Let's get Steve, Spongebob,
35:47 and Mario. So, here's my team. Let's start our battle. I'm playing against
35:54 Caillou, Smiling Book, and Milo Slides. Okay. All right. So each character has
35:59 weapons. So if I select Mario, apparently Mario's got a Glock.
36:01 Mario, a diamond sword, an iron sword, and an RTX5090. All righty then. Uh, let's use the
36:11 Glock, I guess. And, uh, we'll shoot Caillou. And Spongebob just got shot
36:14 with a Glock. All right, let's go ahead and pick Steve. And he can use an RTX
36:21 5090 on Caillou. And now Caillou's dead. Let's try the corridor. I wonder if it's
36:25 just like an infinitely scrolling corridor for no apparent reason. I can't
36:31 like move or change or anything. Like it's just an infinitely going corridor.
36:35 What would you expect when you go to an app called the corridor? And let's try a
36:43 Was not expecting that sound. We've got Mind Sweeper. Does it actually work? Oh
36:47 my gosh. Yeah, Mind Sweeper actually works inside of this Windows XP
36:51 simulator. Nice. There's no way Minecraft actually plays. Oh, it's like
36:55 a fake Minecraft. Okay. And clicking. Oh, we can click to play. I mean, it's a
37:00 fake Minecraft, but it's a pretty convincing Minecraft, honestly. The
37:04 little menu at the bottom looks off cuz it's just colors, but the blocks
37:08 themselves look pretty good. Mario. So, I'm now playing a Mario simulator inside
37:15 of a Windows XP simulator inside of a [Music] Okay, this is pretty epic. I'm not going
37:35 [Music] Anyway, that's what I got for you today. 19 different completely freeto use AI
37:45 tools that do some pretty crazy mind-blowing stuff. Now, I want to
37:48 reiterate what I said at the beginning, though. I have no idea how long these
37:52 will be free. Some of them are just in an alpha phase or a beta phase and
37:56 they're not charging while they're in that phase. Others have like free trials
38:00 where the first month you can play with it for free and after 30 days then they
38:03 want you to upgrade. But I think the biggest advice that I give people who
38:07 ask me about AI and what they should do. I always tell them get out there and
38:11 play with some of the tools. Experiment with them. Find out what you're capable
38:14 of when you try coding. Figure out what you could create when you try these
38:18 image generators. If you see a AI video meme circulating, see if you can
38:21 recreate it yourself with one of these tools. Learn about the tools. Learn
38:25 about what they're capable of. And that's the best way to be informed about
38:29 what's coming and what's available out there and also what sort of superpowers
38:34 you can have using AI as tools in your toolbox. So, I hope you found this video
38:38 helpful. I hope you learned about some new tools. If there's some other free AI
38:42 tools that you think other people should absolutely know about and try, make sure
38:46 you let us know down in the comments. I'm always on the hunt for cool new
38:50 tools, especially ones that you can use for free, and I would love to hear about
38:55 what I don't know about yet. Thanks so much for nerding out with me. I really,
38:58 really appreciate you tuning in and watching these silly videos of me
39:02 playing with AI tools. I really, really hope you're finding them helpful, and I
39:07 will see you in the next video. Bye-bye. Thank you so much for nerding out with
39:10 me today. If you like videos like this, make sure to give it a thumbs up and
39:13 subscribe to this channel. I'll make sure more videos like this show up in
39:16 your YouTube feed. And if you haven't already, check out futuretools.io where
39:21 I share all the coolest AI tools and all the latest AI news. And there's an
39:25 awesome free newsletter. Thanks again. Really appreciate you. See you in the