0:00 Everyone thinks this is completely [music] over, that it's time has passed, that the attention has moved somewhere else, and that the platform behind it doesn't even allow it anymore. But let me tell you something, most people are wrong. One of the jobs of the future was supposed to be creating AI videos and monetizing them on YouTube, or at least that's what most of us thought a few months ago, until YouTube started demonetizing channels that create this kind of content. And most creators abandoned their projects. But if that were really true, why are there still hundreds of channels uploading videos and getting millions of views every month using AI generated content? Because this is not over. The rules [music] have simply changed. YouTube got tired of thousands of people uploading AI slop that adds no value, and that
0:45 [music] is the key. We need to create something that actually provides value. And what is dominating right now and still working is storytelling [music] using 3D animation. So, today I'll show you how, using free AI tools, we can create videos with this animation style, so you can also join this new wave while most people have already left. Better free tools than these could come out tomorrow. [music] If you don't want to miss them, remember to subscribe and leave a like. To start, we need a free text AI tool. You can use ChatGPT, Gemini, or Claude. Any of their free versions will work. And we also need this multi-page master prompt that I'll leave on my website in the description. That's where you'll find all the links for the tools we're using today, the prompts and free [music] text and image AI tools you can use. At the end of the video, I'll show you exactly how to get everything. Once you have the PDF, paste
1:31 [music] the prompt into the text AI, and make sure to modify this field. If you already have an idea for your channel, write it here. The prompt is already set up so [music] the 3D animations come out in the best way possible. It also knows how to create viral content that makes people click and keeps them watching the video. Leave it as it is and send it. What it will do is return different topics you can build the channel around. Now, add the second prompt, and underneath it add the topic together with the information the first prompt gave you. When you send it, it will give you deeper research about that topic and how to make the best possible videos for it. >> [music] >> To increase the chances of the video going viral, I added a lot of the knowledge I've gained from uploading videos for several months. So, it's worth grabbing these free prompts. The part we care about most right now in this report is the final section where it tells us the best videos we could
2:16 make on this topic. Of course, if there's something you don't like, you can ask it to change it. But, if everything looks good, copy the video idea you like the most. Now, add the third prompt and right underneath it, paste the information from the video we just copied from the second prompt. >> [music] >> Here, write how long you want the video to be. And with that, all the technical work is done. The AI has handled it for us. Just send it and what it will return is divided into blocks. Each block will include one voiceover narration line, one prompt to generate an [music] image, and one prompt to generate a video. You'll see that each block is around 5 seconds long, and it will keep giving you numbered blocks in order until the full video is finished. >> [music] >> Now, I'm going to show you the AI tools for images, video, and audio that will turn this into a finished video. And later, we'll also see how to create the
3:02 thumbnails [music] and the channel logo. I'm going to show you the trick most people skip, and that is exactly why their channel ends up looking like AI slop. I don't want the main characters in our videos to keep changing all the time. I want [music] them to stay consistent so we can create things like this. To do that, first open Google Flow and look at the last message the third prompt gave you. At the [music] very top, you'll have a description of what the protagonist of the story looks like. Click new project, but don't generate the images here. Go to the left side and open the characters section. [music] Here, paste the prompt describing what
3:48 the main character looks like. Depending on the story, you may have several characters, but this image generation part is completely free and unlimited. So, you can generate as many as you want. Here, I'm using Nano Banana 2 to generate these images. If the result doesn't convince you, ask it to make some adjustments. The most important part is that it keeps respecting the 3D animation style we asked for from the beginning. Now, if you want this character to appear in all your videos, click the image you like the most, and then click this plus icon at the bottom. It will be added back into this interface with a prompt that appears automatically. If you send it, it will give you a character sheet of the protagonist from different [music] angles, and the platform will save it and use it as a reference whenever you ask. You can also add a voice, so when you turn the character into videos and want dialogue, it always speaks the same way. We'll use that later. Now, create
4:35 the base images that will turn into [music] video. Go back to the text chat and copy the image prompts one by one. Pay close attention here. Next to the number of each block, there is a P icon. That means the protagonist appears in that frame. So, we need to go to the middle panel, because this is where we'll create our images. >> [music] >> This block has the P icon, so click the plus icon, go to characters, and select the character we just created. If you need several characters, select all of them. In this case, we only need one, so leave it selected and generate the image. Sometimes, the P icon won't appear, because what we need is an image of the setting or something that is happening. The protagonist does not need to appear constantly. Generate everything in order. For YouTube, I recommend generating in 16:9, and since this is completely free, generate four
5:22 images at the same time. That way, you can choose the one you like the most. The image quality is really good, and the characters stay consistent. Now, to turn this into video, I'm bringing you three options, so you can choose the one that fits you best. I'll start with the most powerful one. >> [music] >> This method is for people who want to take their project one step further. As you know, Claude, one of the best reasoning models out there, cannot generate images and video by itself. [music] But I managed to change that. I found a method that lets us connect Claude with some of the best image and video [music] models like SeaDance 2.0 and GPT Image 2. To do it, first open Claude. In the menu on the left, click customize, [music] and here select connectors. Now, click the plus icon and choose add custom [music] connector. Give it a name. In this section, open the Higgs
6:07 Field platform where we have some of the best image and video models along with many other tools that we're going to connect to Claude so it can use everything [music] from one place. Go to the MCP and CLI section and choose your platform. For Claude, copy this code here. Then go back to the previous section, paste the code, and click add. >> [music] >> Now, if we go back to a new chat, we can ask it, for example, to generate several images using [music] GPT Image 2 and then transform them into video using SeaDance 2.0. The results are strong enough to use in the actual workflow. [music] I've also been researching it and Higgs Field is currently the most accessible place to use SeaDance 2.0. If you want to try it yourself, I'll leave the link in the description and in the pinned comment. Thanks to Higgs Field for sponsoring this video. If, on the other hand, you want to continue with free
6:54 tools, even if they are not as powerful as the Higgs Field setup, I'm bringing you two more options. Inside Google Flow, you can get some of the best free AI video results available right now. Go to the same section where we were creating the images. >> [music] >> In the option at the bottom right, choose to create videos using ingredients in horizontal format with [music] VO 3.1 light. It says it costs credits, but now we'll see how to handle this. If you click the plus icon, you can upload the images we just created or upload only the protagonist to ask for a shot where the character is doing something. What we'll do is upload the images we created in order. Then go back to the text chat, copy the video prompt, and paste the one that corresponds to each image. Generate them one by one.
7:43 This is currently one of the best free AI video options out there. The problem is that it has a daily generation limit. So, it will only let us create a few videos, and the next day we can keep generating more. That's not bad, but today I found an option that is completely free and unlimited. Open Meta AI. If you go to the create section, you can add the images you want to the flow for free, together with the prompts that correspond to each image. Here, there isn't even a credit section or a subscription section anywhere. And I generated all the videos without any kind of limit. So, if you're looking for a tool to generate infinite videos, >> [music] >> this can work perfectly. Once you have all of them, download them and open CapCut. This is a free editor that works on both mobile and desktop. Add your videos in order, but we're missing
8:30 something very important, the [music] voice. To turn these text lines into audio, open Google TTS Studio. [music] Here, you can choose a voice, copy the dialogue lines in the same order they appear in the text chat, and turn them into one single audio file. The downside is that this tool can get overloaded sometimes, and it may stop letting you generate until some time has passed. >> [music] >> If that happens, I really recommend installing Pinocchio, and inside it installing the Coqui TTS model. If you want to know how to install it, I'll leave the video in the card above. I explain everything there, but you'll see that it's very simple. Once you're inside, this will be a local model, and it will let you use it for free, without limits and without an internet connection, forever. You only need to choose one of these three options. With the first one, you can describe how you want the voice to sound. With the second
9:16 one, you can clone the voice of someone you have permission to use. And with the third one, you can choose a voice from this catalog. Once you've made your decision, download the audio and add it to CapCut. Now, adjust the speed of each video clip so it matches the phrase shown in each block. You can make each clip faster or slower, depending on what the narration needs. Then, I recommend using the transitions and effects that don't have the diamond icon. Those are 100% free in CapCut, and they help add more movement so the video doesn't feel too basic. Now, open Suno, go to the create section, and choose [music] simple. If you look at the prompt from the blocks, right under the description of what the main character looks like, there is a line of text describing the perfect background song for this video. Copy that line and paste it into simple. Click generate, and it will create a
10:02 song that works perfectly as background music. >> [music] >> The dust storm was already swallowing the dead city, and his last air filter was breaking. Only one building still had emergency medical supplies. If he found a working gas mask, he could survive. >> Finally, [music] to create the channel logo and each thumbnail, the previous prompts are already designed to know what they need to do. Ask the text AI to give you a prompt for Google Flow based on the last video, describing what the perfect thumbnail would be. It is already designed to be as eye-catching as possible, and to stand out in users' feeds. So, just paste that prompt into Google Flow, choose the image model, and add your protagonist as an element. The result is a really strong thumbnail. Now, go back to the text chat and ask it to create a prompt for the perfect
10:48 channel logo, taking into account what your videos are going to be about. It will give you a prompt, and if you paste it into Google Flow, this time choosing this image format, it will generate the logo for your channel. Remember, these image tools are unlimited and free. [music] If you don't like the results, keep generating, or add one as a base image and ask it to make some changes without completely destroying the image. With this, you have the full workflow to start building the project. To get the prompts, the direct links, and access [music] to the free AI tools, go to my website in the description. Open PDF guides, and you'll find the guides for the videos I have published so far. You can download them and access the information from each video, so you can copy the prompts and links without wasting time writing everything manually. You'll also find the free AI chat there. You can ask questions
11:34 without paying, so it is useful if you want a free chat option for prompts, questions, or quick tests. We also added a model that generates very high-quality images. You only need to sign in and start using it. See you in the next one, and thanks for watching. [music]
