This is the full transcript of Claude Fable 5 + Higgsfield AI = $60K/Month Faceless AI Channel (2026), published on YouTube by Sanji Nai-Chien. Every paragraph carries the moment it was spoken, so you can click any line to jump straight to that point in the video, search the whole thing for a word, or copy it out.
0:00I just found one of the fastest growing AI YouTube channels I've ever seen and it's making over $60,000 a month. Now, I wanted to see if I could recreate it in just 20 minutes. [music] The crazy part is this channel only started 3 months ago and it's already practically printing [music] money. So, if you want to learn how to use AI and AI automation to build a YouTube channel like this and start your own YouTube journey, this video is for you. I'm going to show you the entire workflow step by step. So, let's get started. So, here's the channel. It's called Zen and just take a
0:31look at these videos. They only started uploading about 3 months ago, but here's what you really need to pay attention to. This channel has already generated over 16 million views and if you click on the VidIQ stats, you'll see it's estimated to be making over 60k a month. Now, if you've been wondering whether YouTube blocks AI generated content, the answer is no. So, yes, YouTube absolutely does monetize AI generated videos. It all ultimately comes down to the quality of the content that you create. Now, if you look at the channel's most popular videos, you'll notice that this format gets an absolutely crazy amount of views. So,
1:06I'm going to show you exactly how to recreate this entire channel with AI in about 20 minutes and finally start building your own YouTube business. Now, for this to work, you'll first need to download Cloud Code on your computer. Just head over to Cloud, download Cloud Code for your operating system, install it and you're ready to go. Once Cloud
1:25Code is installed, go ahead and open it. Now, before we can generate anything, the first thing we need to do is connect Cloud Code with Higgsfield. So, head over to Higgsfield, create an account if you haven't already and click on MCP and CLI. Now, once you're there, you'll see your MCP connector. All I want you to do is copy it. Now, go back to Cloud Code, click on your profile, open settings, go to connectors, click add custom
1:49connector and give it any name you want. I usually just call mine Higgsfield. Then, paste the MCP connector you copied from Higgsfield. Click save, and that's literally it. Cloud Code is now connected directly to Higgsfield, which means every time Cloud Code needs to generate an image, it'll automatically use Higgsfield in the background. There's no complicated setup. You
2:09connect it one time, and you're done. Now that everything is connected, we can actually start creating the videos. Now, what you need first is the YouTube script. Depending what kind of video you want to make, maybe you want to talk about history, space, aliens, animals, it really doesn't matter. Once you have a workflow like this, you can apply it to almost any niche on YouTube. Now, how you create the script is entirely up to you. Now, once you have that script,
2:34you'll [music] need the voiceover. Personally, I use AI voices because they're much faster when you're creating videos like this, but if you prefer using your own voice, that's completely fine, too. Now, once you have your voiceover, I want you to go back to your browser and go ahead and search for Turbo [music] Script. Go to the site, create an account, and click transcribe files. Upload your voiceover, and then click transcribe. Now, you might be wondering, why are we doing this instead of simply sending the script directly to Cloud Code? Well, we're not actually here for the transcript. We're here for the timestamps. When the transcription is finished, you'll notice those
3:10timestamps, like 0 seconds, 7, 15, 23, and so on. And those timestamps are the most important part of this workflow. Because instead of manually deciding where every image should appear, we're going to let Cloud Code handle all of that automatically. Every timestamp becomes its own image, which [music] means Cloud Code already knows exactly where every illustration is going to belong in the final video. Now that we have the timestamps, we can finally start generating the images. Go back to Cloud Code and create a new chat. The first thing we're going to paste is our master prompt. Now, you can think of this prompt as the director for the
3:47entire video. Instead of asking Cloud Code to generate one random image at a time, we're giving it a complete set of instructions for the entire project. We're telling it to read the entire script first to understand what's happening in every single scene. Generate one image for every timestamp and make sure every generated image matches the exact moment where it appears in the narration. The second part of this prompt is just as important. This is where we define the visual style. Now, in our case, we want pretty simple illustrations, MS Paint drawings, beginner hand-drawn style with and the reason we're doing this is because that's exactly what made this
4:24channel successful in the first place. The viewer isn't watching because the drawings are beautiful, they're watching because the story moves fast. The visuals simply help tell that story. Now, underneath the prompt, paste your transcript with all of the timestamps included. Don't paste the original script. Make sure you're using the timestamped version from Turbo Scribe. Once everything is ready, send the message. Claude Code will automatically read the transcript, understand every timestamp, and start generating images through Higgsfield. If your script has 40 timestamps, you'll get around 40 images. If it has 80 timestamps, you'll get 80 images. One timestamp, one image across the entire video. Once all the
5:04images are finished generating, the next thing we want to do is save everything locally. Just ask Claude Code to download every generated image into a folder on your computer. And while it's doing that, ask it to rename every image using its timestamp. So, instead of random file names, you'll end up with files named 0 seconds, 7 seconds, 15 seconds, and so on. It might sound like a small detail, but this is what actually makes the editing process incredibly fast. Now, open your favorite video editor and import your voiceover first. Then, drag in the entire folder with all the generated images. Because every image is already named using its
5:39timestamp, you're going to immediately know exactly where it belongs on the timeline. The 0 second image starts at the beginning, it's then replaced with the 7-second image at 7 seconds, and so on and so forth. And you simply keep going until the end of the video. You're never going to have to guess where every image should appear, and you'll never have to watch the voiceover over and
5:59over again to try to sync everything. Cloud Code already did that for you. Now, at this point, AI has handled almost the entire production process. All that's left is arranging the images on the timeline and exporting the final video. The editing itself takes around, I don't know, 10 minutes. And once you're happy with the results, export the video, upload it to YouTube, and start working on the next one. Because the real advantage of this workflow isn't that it helps you make one video faster, it's that it helps you develop a system that you can repeat over and over again. Now, let's compare the final result. On the left, we have the
6:35original channel. On the right, we have the video we just created. The style is almost identical. Simple hand-drawn illustrations, fast pacing, and the entire workflow was powered by Cloud Code and Hexfield working together. What's even more impressive is that we didn't spend hours writing prompts for every single image. [music] We generated the entire sequence from just one script, one voiceover, and one master prompt with a single workflow. That's what makes the process so powerful. Now, does this mean you're going to start making 60k a month tomorrow? Of course not. That channel didn't grow because of AI, it grew because it consistently published content that people actually
7:14wanted to watch. AI simply makes the production process dramatically faster. The people who win won't necessarily be the ones with the biggest teams, they'll be the ones who built the best systems. So now, it's your turn. The link below has everything you need. Open Cloud Code, connect Hexfield, build your first workflow, and publish your first video. Because the sooner you start, the sooner the YouTube algorithm starts pushing you. And that's the only way to figure out what actually works. I'll see you
7:44guys in the next one.
No line in this video contains that word.
Where these words come from. This is the caption track YouTube holds for this video, written automatically by YouTube rather than by the creator. We read it, tidied the line breaks and laid it out so it can be read. The plain text version is at https://viewrankai.com/tools/youtube-transcript/f-ufEhGtVpw.txt.
All rights in this video belong to Sanji Nai-Chien. Watch it on YouTube. If this is your video and you would rather this page did not exist, tell us and we will remove it.