Manus AI Agents + Veo 3 = Automated YouTube Empire 📈

Manus AI agents now have Veo3 built right in. And this is your blueprint for creating an automated YouTube video empire that runs itself. I'm walking through the secret behind those viral Yeti videos hitting 2. 1 million views. My complete workflow setup that takes you from zero to a full content library without touching a camera or microphone and the exact costs, time frames, and technical setup that is required to copy this system step by step starting today.

So, Manis have a brand new update where you can basically integrate Veo3 generate AI videos like you can see right here. I'll play this in a second. And essentially with this, you can now generate longer clips. You can tell the AI agent to generate videos, and they're very, very lifelike. So, let's play back this video, and I'll show you exactly what we mean and how to use this in a minute. The miracle begins. Now, Manis is here. What comes next is V3 inside Manis. Let me show you. Every director has a language. Could AI learn to speak those languages? So I gave Manis a simple prompt create. So essentially if you want to start using this, right, you can go into Manis and then if you want to see an example of what we did here. Well

basically we can generate very cinematic videos using Manis and it will generate these clips and even embed them directly into a slideshow or into a presentation as we can see right here. So for example, what I did was I said use VO3 generate a series of short cinematic videos each in the distinct visual and storytelling style of a legendary film director should be like a film style etc. And then if we actually open this up we can play the videos. So, this is the Alfred Hitchcock video. Let's play this back. So, it's got like that cinematic style. Let's open up the one on Wes Anderson. I think I'm generated by prompts. That explains your quirky dialogue. We've got Christopher Nolan style here. Am I real or just generated text? This is pretty cool. I really like the neon colors in this. Are my emotions programmed?

All right. And if you're not aware, some people are aware of what Manis is, some people aren't. So Manis is a super agent that can build tools, websites, games. It can research trips for you. It's basically like an AI assistant that can do a lot of things that an actual employee can do um with like a kind of chat GPT style interface, but the biggest difference is here that it can also generate videos as well. And it's kind of like a wrapper for other API keys. So for example, like it can use Claude and Bayer 3 and everything else. So let's play this one back as well. Am I just following a script? pretty nice video generations. That's Tarantino style apparently. And then this is like a kind of fantasy style. Maybe. So you can see the power of this. What you can actually do as well, I think, is you can get Manis to like stitch up all the videos and then create like a minute long video or even longer. And if we go inside the chat, this is exactly how I generated the videos. So, I use this prompt right here. If you want to steal the prompts from me, they're inside the AI success lab along with a bunch of training links to different courses on this stuff. And you can see that I've included links to examples, plus the copy and pasteable prompt if you want to just customize that to you. So, if you wanted to get started with this, what we can do is we can go over to Manis and then we'll click get started in the top right. I'm on I'm trying to think which plan I'm on. Yeah, I'm on Manus Plus here, which I think let's see how much it costs. Yeah. So, these are the plans, right? So, I'm on the $39 a month plan here to generate these videos. And then if we go inside here, I'm going to say use this to create the content. And I'm going to say make sure you give me the prompts to approve first. I'm going to say it must be vlogging style. Keep all the prompts. scene descriptions as vlogging style and keep the dialogue short as Veo3 only allows I think it's maximum like 8 seconds but click right so we're going to go with that first we'll hit enter now I've

used other tools as well. Like you've got other agents that can do this like Gen Spark as well. Honestly, if I have a choice, I'm probably I'll show you the outputs in a second, but I would say I'd still prefer Gen Spark, but I want to show you cuz I know a lot of people here use Manis and that sort of thing. I just find Manis is a little bit slower than Gen Spark when it comes to generating outputs. But still, it's amazing that you can do this with either one. And you can see the process here, right? So it's like right minus is going to think about this first then we're going to get user approval. generate the videos and then we're going to deliver the final videos to the user. Right? So it's generating the outputs here.

Now you might be thinking okay isn't this just like AI slop or nobody's going to watch this etc. So what you can actually do is if you go on YouTube and actually Tik Tok as well is a good example of this right? So if you go on Tik Tok, you can see here like some of these videos are absolutely blowing up, right? So for example, this account 2. 1 million views, 1. 2 million views, 320,000 views, right? Like these Yeti style videos are absolutely blowing up um and performing very well on social media. So you can create viral videos. They do get a lot of views um

and they perform very well. So let's have a look at the prompts to approve here. So, morning routine introduction. A massive fluffy white yeti holds a smartphone and selfie stick. Speaking directly to the camera in a vlogging style, blah blah. Says, "Good morning, Snow Squad. Welcome back to my channel. " All right, so this looks good to be honest. So, I'm going to say approved now. And then from here, we can wait for Manis to start generating the videos. And it's doing its magic right here. Now, you can see here it says to get better video quality, Manis recommends using the new Veiler 3 model. It costs 300 credits for 8second clip. Um, it's only in that dimensions, the 16 to9 format only. The default model costs 30 credits. I'm going to continue with A3. Ve3 is the model you want to use, right? If you use something like, for example, um, Cling 2 or something like that, you're not going to get the audio dialogue synced with the video. So, you want to make sure you're using Bay 3. Tyler is asking his appearance change throughout each clip. No way to control that. Yeah, that's very hard to control. you can have like I think what Manis did for example in that scene that we looked at before to get continuity is they probably used the same description of the scientist in each scene. So like this one it looks very similar to this one cuz he's got glasses and a suit on. Um but yeah, that's the way that I see it. I think like that's the only way to get continuity. But you know if it's something generic like Bigfoot, most people are not going to notice a difference to be honest with you. So you can see now it's generating the video and it's a lot slower than using something like Manis uh sorry something like GenSpark. So when you're using Manis for example you can see it's just generating one video at a time. It's not generating like all 10 videos at once right and so you have to wait for each video to be generated before that's going to be done. Now let me show you how Gen Spark does this in contrast. So, if we have a look at Gen Spark, which is another AI agent that can do the same thing, and we'll pull these up side by side. So, what Gen Spark actually does is it generates all the videos at once side by side and it's a lot faster, right? So, for example, Manis is just creating one video at a time whereas Gen Spark is creating all the videos at once, right? And I much prefer that personally just because I think you get faster and better outputs. And for me, I'm always trying to optimize for time, right? That's the whole point of doing this. So, if I have a choice, I'm still going to go with Manis, uh, sorry, with Gen Spark, but Manis is doing okay. Here's another bunch of videos we actually created with Gen Spark, which are pretty cool, nice and colorful, etc. as well. And then we can wait for the outputs to get back here. All right. Now, whilst we're waiting, you might be wondering, okay, like is Manis just like an AI video tool? Absolutely not. So, if we look at some examples here, let's pull up Manis. So, if we pull up some examples of Manis, like you can build other stuff like for example 3D runner games. Like you can see, you can create SEO tools. This is another SAS tool we create for AI SEO content writer. Actually takes an API key and then you can plug in the API key and generate content with it. We've created a full SEO directory using um Manus and also like it can do a lot of deep research and then plug that into a permanent URL. Additionally, here's another website that we created and also like a kind of Mario style game as you can see where you can just run around and play like Mario basically. So, it can do a lot of other cool stuff as well, but today we're focusing on the Manis scene creation. So, now it's generated that video. Let's play this back. We can actually download that. Takes ages to download to be fair. Let's try and play the video back. So, we've got the video back. Now, this is the first scene from the Yeti vlogging. Looks pretty cool already. Like, the visual is super nice. You got the little selfie stick right there. Good morning, Snow Squad. So, it looks super cool. Really easy to generate that. If we X off that, we'll go back to the chat. Now, it's generating the next video, right? And so, it does take quite a long time to generate these videos. morning. Step by step. What I would recommend is you just go off and do something else whilst this generates the outputs for you. Otherwise, you're going to be waiting absolutely ages. I mean, you can see how this if you were creating eight scenes. Well, that's what that's 8 time 3 minutes. That's 24 minutes of just waiting around for the videos to generate. So, it's going to take ages. Let's click on the next video and see what that's like. So today, so we got the next scene here. You can see as well like it's not quite that continuous, right? So it's it just looks slightly different versus the first scene, but it's still cool. So today we're making my famous Yeti latte and it's going great. All right. And then it's just generating the video step by step right there. Now, once that's done, obviously I'm not going to make a 24-minute video because that'd be very boring for you to just wait for Manis to be delivered. But once it's generated all the videos, then what you can do is you can actually I think you can ask Manis to stitch all the videos together. And from there you can create like a minute long video or you know if you created 80 of these well that's let's have a look 8 * 80 cuz each video clip is 8 seconds. So that's 640 seconds. So that's about 10. 5 minutes long right? So, if you got 80 clips from Manis, which would use up a lot of credits, but if you got 80 clips, then you would actually have like a full YouTube video vlog ready to go. And literally all you did was just ask Manis to do it, and then you go off, you wait for it to load. Maybe it would take about an hour to generate all of those and you're good to go, right? Pretty wild. So, that's how you can create all these videos. Like I said, I still prefer GenSpark. like GenSpark's process is a lot faster and easier to be honest, but it's amazing that you can do this with any platform. Um, and it's crazy how powerful this is. You can also take the API and you can use something like foul and a combination of foul plus uh Veo3 to generate outputs as well, right? So, for example, if we go to

explore, you can get Veo 3 over here. And then you can actually plug this into N810 and generate like a sort of one-click workflow where you can generate outputs. So, let me show you an example of that. So, for example, if we have a look, this is a BO3 NA10 automation that I built. You can see how it works. Basically, what it's doing is it's grabbing some ideas. It just comes up with random ideas. Then it uses the AI agent to come up with the script. Then it plugs the prompts into Veo3's API on foul. That generates the output and then that goes to a Google sheet with a breakdown of all the videos we've created. And that's like in a importable NA10 workflow that you can just download from me inside the AR profit boardroom and then upload into NA10. Right? If you want to see some examples of that in action, here's a spreadsheet with all the files. Let's open this up and see what we get. How do Can you believe this? They're actually dancing. So, this was fully automated with NHSN. Check it out, folks. Totally organic wild squirrel smoothie straight from my backyard. Here's another one. Right, that was another one. That's Yeti losing his selfie stick in the snow. And then we have Yeti eating cereal with chopsticks for some reason. How do people even do this? Right? And you can automate all of this with uh workflow on NA10, right? Pretty simple

using the AI profit boardroom. We also got, you know, if you want to learn like how to build this stuff and how to customize it yourself and that sort of thing inside the AI profit board and we do like three weekly coaching calls. Plus, you can ask any questions inside the community in between the coaching calls. If you can't make the calls, we actually record them and you can watch them back. And we have a full NA10 templates and workflow section here with loads of different NA10 automations, right? So, for example, like this is an NA10 automation for generating avatar videos with like one click. We've also got a firecrawl like Reddit AI agent that can research the internet and you can just download these templates, upload them into NA10 and then you're good to go, right? And additionally, what we also do each week is we just ask you like what do you want to see? And then from there, you can just post whatever you want to see. We'll create the automation for you and we'll create a video tutorial on exactly how to do it. Right? Someone is asking like how do you keep the characters consistent? I've already talked about that a little bit, but basically with the character consistency, it is tough. It is tough, right? I think the best way that I've done it so far with GenSpark and just asked it to keep like one consistent style throughout the content, right? So, for example, if we go back here, go into my recent chats, this is a Bigfoot jungle workout, right? And you can see here that it all kind of looks very similar, right? So, for example, you got Bigfoot here and and here and also it's all like the similar style, right? One of the easiest ways to do this is you just tell it exactly what you want. So, for example, you can say, right, make it feel and look like a YouTube vlog and you do that throughout the content consistently and then all the videos have this character consistency like you can see right here. And that seems to work pretty well. It's not perfect, but it does the job. So, if we go back to Manis, I think it's generated four videos now. Let's open up this one and we'll see what this is like. So, we got the Yeti skincare routine. Step three of my skincare routine. Organic mountain snow. And then we have video four here, too. Time for my daily yoga flow. Mountain pose is literally perfect for me. Love that. Like it's just really uh the visuals are really cool. And then you can see it's running in the background whilst I've been running through, you know, all the other stuff on today. So yeah, that's basically it. Thought says, can it make reals Tik Tok or only landscape? So you can actually post Tik Tok uh landscape and it works really well, right? So for example, I don't know if you ever seen this, but the Yeti cartel is a good example of this. So you can see here it's a landscape video. It's absolutely awesome. This is Yeti Cartel on Tik Tok. And you can see it's a landscape video, but it works really well. And uh that's a minute four collection. So it's basically like a it's a full video generated and then stitched together. Like you can see it looks really cool. The visuals are absolutely amazing. And then you know you don't need to worry about creating portrait videos. But yeah, BA3 with its API only creates landscape, but you know on Tik Tok wherever you can post landscape. So that's not a big issue. And then if you want to get a reals automation, you know, where you generate like portrait videos inside the inside here, we have the NA10 viral reels workflow. You can download the JSON here. So you just download this shorts automation and then upload it into NA10 and you're good to go on that. it will create like a nice little um portrait automation, but that's using Cling 2. 1 and Flux. So, you're not going to get the audio on that. And that's why I recommend the landscape version instead.

All right. So, thanks so much for watching. I've shown you exactly how to use Malice, the whole process right here. Hope to see you in the air profit boardroom. And if you want us to just get all of this stuff done for you, if you like Julian, I need this stuff for my business. I don't know how to implement it. I just want to hire someone else to do it. I don't want to figure it out myself. then feel free to book in a free AI strategy session link in the comments description. We'll basically jump on that call with you. We'll look at your business. We'll look at what you want to automate and where you're spending all your time, right? And then we'll go, right, let's focus on building an automation for you each month that saves you the 8020 of your time, right? So most business owners, they're spending most of their time on one or two things that can be easily automated. And that's what we focus on. This is how I've scaled my own business. It's just finding that one or two automations each month that I can automate and then from there go on. Um, someone else asked, does it have copyright restrictions? I'm not a legal expert, so I'm not going to give any legal expert advice on that. I would just ask Chat GPT or, you know, if you're really worried about it, ask a legal expert. I'm probably not the right person to ask on that. All right, so thanks so much for watching. Appreciate it. I will see you on the next one. Cheers. Bye-bye.