Big Idea

The argument in one line.

YouTube's semantic ID punishes copied channels, so the winning move is to take a proven format and change the character or packaging until it attracts a different audience, then use Claude Code to produce it fast.

Who This Is For

Read if. Skip if.

READ IF YOU ARE…

You want to start a faceless YouTube channel and keep seeing the same stick-figure or animation niches that everyone is already copying.
You already copied a successful channel with new colors and can't figure out why it isn't growing.
You use Claude Code and want to see a concrete workflow that turns image generation into a repeatable animated-video pipeline.

SKIP IF…

You want proof of a repeatable income result: the revenue shown is from one channel over three weeks, and he says plainly that many people in his community don't make it.
You need a deep technical build guide: the Claude Code and Higgsfield section is a short overview with one example prompt, not a step-by-step.

TL;DR

The full version, fast.

Danny Why shows a Shorts channel he built with Claude Code that earned almost $2,000 in three weeks and drew about 50 million views. He argues that YouTube's semantic ID flags videos with similar titles, thumbnails and scripts as inauthentic, so copying a channel fails. The fix is to inspire from a proven channel and change one element, like swapping a stick figure for an anime character, so the audience differs. His example is a Peter Griffin workout channel built by connecting Higgsfield to Claude Code, which writes prompts, removes backgrounds and animates PNG characters into a finished video in 20 to 30 minutes.

Free for members

Chat with this breakdown — free.

Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.

Create a free account →

Chapters

Where the time goes.

00:00 – 02:00

01 · Proof: the channel's analytics

Shows about $2,000 in revenue over three weeks, 50 million total views, and why 35 million of them are unmonetized.

02:00 – 03:52

02 · Disclaimer and the oversaturation problem

Results vary, YouTube is saturated because AI lets everyone publish, and the advice is to start immediately.

03:52 – 05:30

03 · Semantic ID and why copying fails

Explains YouTube's semantic ID: similar titles, thumbnails and scripts get classed as inauthentic.

05:30 – 08:20

04 · Inspire, don't copy: Yellow Dude vs the anime channel

Compares a 2 million subscriber stick figure channel with an anime-character channel that hit 7 million views on one video.

08:20 – 10:40

05 · Why a different audience reads as original

The algorithm sees football and hockey fans on one channel and anime viewers on the other, so the second counts as new.

10:40 – 11:36

06 · Combine niches: the Nick Finance example

A Family Guy character teaching finance shows how to borrow a mascot from one niche for another.

11:36 – 13:30

07 · Tools: Higgsfield and Claude Code

Why Higgsfield is the image tool and how to connect it to Claude Code through MCP.

13:30 – 15:40

08 · The workflow and the result

Prompt Claude for a muscular Peter Griffin workout channel, get thumbnails and titles, then an eight minute animated script in 20 to 30 minutes.

Atomic Insights

Lines worth screenshotting.

Copying a successful channel with new colors fails because YouTube's semantic ID compares titles, thumbnails, scripts and audience, and flags near-duplicates as inauthentic content.
Changing one packaging element, such as the character, can change who clicks, and a different audience is the strongest originality signal the algorithm sees.
Two channels can teach the same topic and still be treated as original if one draws football fans and the other draws anime fans.
A channel with only 11 videos and 60,000 subscribers can have a 7 million view video, so a fresh twist on a saturated niche still gets pushed.
Shorts views don't earn money until you pass 20 million views in 90 days and get accepted to the Partner Program, so most early views go unmonetized.
Take a mascot from a popular cartoon and put it in an unrelated niche, as Nick Finance did with a Family Guy character teaching finance.
Claude Code writes more detailed image prompts than a person does, and it can produce 20 of them in about a minute.
Higgsfield's background remover, called through the Claude connection, turns generated characters into transparent PNGs that Claude Code can then animate.
A posed character reused across many generated frames makes a 2D animation read as hand-made rather than AI-made.
The workflow is copyable. The channels, niches and packaging aren't, and that is the dividing line the whole video draws.
Danny's own caveat: it worked for him, it won't work for everyone, and some members of his community never get monetized.

Takeaway

Change the audience, not just the colors

WHAT TO LEARN

A new channel survives saturation by borrowing a proven topic and changing the character or packaging so a different audience shows up.

YouTube compares titles, thumbnails, scripts and audience, so a near-copy of a channel gets labeled inauthentic and isn't pushed.
Changing one visible element, such as an anime character instead of a stick figure, can pull in an entirely different group of viewers.
A different audience is the strongest originality signal, more than any single design tweak.
Shorts views don't pay until you clear 20 million views in 90 days and are accepted, so plan for a long unmonetized stretch.
Borrowing a mascot from one niche to teach another is a repeatable way to build a distinct channel on a familiar topic.
Claude Code can write detailed image prompts in bulk and, with an image tool connected, turn them into posed, background-free characters.
Reusing one posed character across many frames makes the animation read as hand-made, which helps retention and credibility.
Treat the results as one data point: the creator admits some people in his community never get monetized.

Glossary

Terms worth knowing.

Semantic ID: The claim in the video that YouTube's AI algorithm understands what is inside a video, and flags ones too similar in title, thumbnail and script to an existing video as unoriginal.
Faceless channel: A YouTube channel with no on-camera host, usually built from animation, stock footage or AI-generated visuals and a voiceover.
YouTube Partner Program: YouTube's monetization program. For Shorts, the video says you need 20 million views in 90 days to apply, then wait days or weeks for approval.
MCP: A connection standard that lets Claude Code call outside tools. Here it links Higgsfield's image tools so Claude can generate and edit images directly.
Packaging: The title, thumbnail, character design and visual style that decide who clicks on a video.

Resources

Things they pointed at.

11:40productHiggsfield ↗

12:10toolClaude Code

05:50channelYellow Dude (YouTube channel)

10:40channelNick Finance (YouTube channel)

15:30linkThe Creator School (Skool community) ↗

Quotables

Lines you could clip.

05:40

“The secret is to inspire from them, not to copy them.”

Short, contrarian, no setup needed.→ TikTok hook↗ Tweet quote

03:55

“Because of AI, everyone can make videos now. So if you want to make this work, you need to start as soon as possible.”

Urgency framing that stands alone.→ IG reel cold open↗ Tweet quote

07:40

“Even changing one small part of your video can completely change your audience.”

One clear, transferable rule.→ newsletter pull-quote↗ Tweet quote

The Script

Word for word.

Read-along

Don't just watch it. Burn it in.

See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.

Okay, so this is the channel I've recently started using cloud code. Here are the analytics. And if I click over here on revenue, you'll see that I made almost $2 ,000 in the last three weeks, almost three weeks.

If I select over here, the last 365 days, you will see that I started earning literally like two to three weeks ago. Same for subscribers. I started getting subscribers like three weeks ago, watch time and views.

And you can see I made almost 50 million views in total. let's go back to the last 28 days so you can see the channel is earning around 100 to 200 dollars per day it's been going down a little bit because i didn't have enough time to work on it but this is a channel and i started it using cloud code now Something you need to understand is that 35 million views from this channel are not monetized.

That's because on Shorts, in order to get monetized, you need to reach 20 million views within 90 days in order to apply to the Partner Program. And then once you apply, it takes a couple of days or weeks until you actually get accepted into the Partner Program. So basically, a lot of those views are not monetized.

If they were, this channel would probably already earn me $6 ,000 to $7 ,000. now you're probably asking yourself what videos am i making and how am i making them well i'm making minecraft animations and i'm using cloud code to help me create these videos now in this video i'm going to show you how you can make a channel like this one too i'm going to explain how it works why it works whether it's worth it for you to try and how much it can cost you now a quick disclaimer just because it worked for me doesn't mean it's going to work for you right Now, I do have a private community and because I work one -on -one with everyone who joins, people in there do get monetized.

For example, this person right here got more than 80 ,000 views on his first upload and he's going to get monetized really soon. But there are also people who don't make it. That's just how the world works.

And I need you to understand that before you watch this video. So please keep that in mind. With that said, if you want to try to grow your own YouTube channel and make money with it, this video is for you.

Let me show you the workflow. Now, first of all, you have to understand that YouTube is quite oversaturated lately. Because of AI, everyone can make videos now.

So more and more people are making videos. So if you want to make this work, you need to start as soon as possible. I'm literally asking you to start today or tomorrow or just as soon as possible.

And you also need to understand that almost every single niche is getting oversaturated too. Why? Well, because people are learning more and more.

There are communities like mine where I help people learn how to package and how to do storytelling. And I've been doing YouTube since 2009. And my community is not the only one.

There are more communities teaching people how to do this. Because of this, more and more regular people are learning how to make YouTube videos properly, how to package them, how to do storytelling, how to keep retention, how to get people to click. And because of this, the platform YouTube is becoming more and more saturated.

So how do you win? How do you grow a brand new YouTube channel when the platform is getting oversaturated? Is it even worth it to spend time creating a brand new channel?

Well, I've just shown you my channel in the beginning and it is worth it. You can literally monetize it within weeks if you know what you're doing. And that's exactly what I'm about to teach you inside this video.

And I'm not going to go and tell you to go ahead and copy someone else's channel because it doesn't work like that. People who tell you to do so... They're lying to you and they're just trying to get your money.

What I'm about to teach you inside this video is how to use your brain to create your own niche that people are interested in. And I'm going to show you the workflow I use inside Cloud Code to create these videos within 10 to 20 minutes, because this is something you can copy. You can copy this workflow, but you can't copy other channels, niches or topics or packaging.

And let me explain why. Now, first of all, I'm struggling to breathe. So I'm going to add one of those patches to my nose.

A lot of you have been asking why do I use these things sometimes in my videos and why some other YouTubers are using this. By the way, I'm not sponsored. I'm not even showing you the brand.

The only reason why I'm using them is because they open up your nose. So you can see over here, if I put them, they stick to my nose. They open up my nose and I breathe a little bit better and I'm able to talk faster and more cleaner.

So here's what you need to understand. YouTube has something called semantic ID. It's part of their new AI algorithm.

And it basically understands everything that's in your video. If your video is too similar to another video, it basically has a similar title, a similar thumbnail, a similar script, the semantic ID within this algorithm will detect your video and will categorize it as inauthentic content, not original content, because there's already a video like that that exists.

And because of this, YouTube is not going to push your video. Why? Because YouTube wants originality.

That's exactly why YouTube does this. YouTube does this to preserve originality on the platform. Because if every creator copies each other, the platform becomes unoriginal.

Every content will look the same. And if every content looks the same, the platform becomes boring. And if the platform becomes boring, viewers leave.

Right? Does it make sense? I think it makes sense, right?

That's why I'm saying you can't copy other channels. It doesn't work. It used to work two to three years ago, but now it doesn't work.

Now you have to actually be original. So let me help you out. Let me show you how I get inspired from other channels to create my own niche.

Look at this channel over here. It's called the Yellow Dew. It uses this kind of 2D animated stick figure to show you how to build muscle.

it has almost 2 million subscribers and it's getting millions of views it probably made even over 1 million dollars so far it's probably even selling digital products or a digital course or whatever now a lot of people saw this channel and started copying it i even got messages inside my private community from members showing me their channels and their channel was literally a copy of this channel the only thing they did was change the colors or a little bit of the design but it was very easy to tell that they were copying this channel And they were asking me inside my community, why isn't my channel growing?

Well, I told them, look, your channel looks almost identical to this other channel, you know? It's kind of obvious. But that doesn't mean it doesn't work.

I've helped people inside my community get monetized within three weeks. Here's a testimonial from Eugene. And basically all I did was help him change his packaging, which in the end resulted into changing his niche, into making his own original niche.

And I did this by inspiring from the channels they were copying. Because the secret is to inspire from them, not to copy them. So let me show you how inspiring works.

Let me give you an example. Look at this channel over here. Right now, it has around 60 ,000 subscribers with only 11 videos uploaded.

And if we look at the most popular uploads, we will see that one upload that was posted just two months ago got over 7 million views. Now, based on my own experience of running faceless YouTube channels and running Denny Why and mentoring people, 7 million views on a channel like this usually earns you around $10 ,000 to $20 ,000, which is pretty crazy from just one video.

Now, do you notice something? Do you notice that this channel is quite similar to the Yellow Dude channel, which has 2 million subscribers, right? But the difference is that this channel has completely different packaging.

It uses an anime character, which means the people clicking on these videos are completely different from the people clicking on the Yellow Dude's video. so you need to understand that even changing one small part of your video can completely change your audience for example the yellow dude's stick figure looks like this so the channel is probably attracting people who like football hockey people who are more into social stuff or whatever right but this new channel over here on the other hand uses an anime character so it's probably attracting younger teens who like to watch anime or play video games so these two channels over here are essentially talking about the same thing but they are attracting different audiences which makes them different because the algorithm will see that this channel over here called yellow dude is attracting an audience of football players and hockey players i don't know a completely different audience probably people who don't watch anime and this channel over here is attracting an audience of people who actually watch anime and they don't watch football and because of this

This new channel over here, which uses an anime character, even though it's basically inspiring from the yellow dude, it feels completely new and original because the audience is completely different from the yellow's dude audience. And the packaging on this new anime bodybuilding channel is also a little bit different. It's black and white.

which makes the channel feel much different from this one over here, even though essentially, as I said, they're the same thing. Believe it or not, the YouTube algorithm notices this. Semantic IDs sees that the yellow dude character has a very specific look, and the character on this anime channel has a different look.

But not only that, the algorithm can see that they have different audiences and that's the strongest signal that your channel is original. You have a completely different audience from this channel over here, even though you talk about the same thing. Basically, when you copy a channel, let's say you make your own yellow dude channel with its same kind of animation, YouTube can tell right away that you're copying it because usually you end up with a very similar audience.

and then it connects the script and the thumbnail and everything and then your channel is cooked. It just dies. But when you make a channel like this one with an anime character, you attract a different group of people.

People who usually watch anime and they don't watch the Yellows Dude channel. And so the YouTube algorithm notices this and goes, wow, this is completely original. This is an original channel.

I have never seen the anime audience watch a workout channel. Let's push this channel. And that's exactly what is happening right here.

That's how it works at its core. And that's exactly what you have to understand. You can't just copy channels.

You need to look at them and see if you can find a way to bend them, to twist them, to create your own niche and attract a new audience by making essentially the same videos, right? But you're probably asking yourself, is there any space left for me like the yellow dude channel already exists and has almost 2 million subscribers this channel over here already exists and has 60 000 subscribers how can i create my own niche is there any space left well that's what i'm about to help you with i'm gonna show you how I would build my own bodybuilding channel if I wanted to build a faceless channel like this one.

So let me explain. The idea is super simple. You basically take multiple niches and create your own niche by making basically the same videos.

So let me show you. there's this other faces channel called nick finance which uses the family guy character to talk about finance so what this channel nick finance did was basically exactly what i'm teaching you instead of just making generic finance videos like everyone else he took a character like he took something from a different niche and talked about finance which is basically the same thing other channels talk about but because he used the family guy character he attracted a completely different audience which made his channel feel original And basically, we can do exactly the same thing.

We're going to take a character from a popular cartoon, use it as our mascot, and teach people how to work out. So for example, you can take Peter Griffin from The Family Guy, give him muscles, and then create your own workout channel with Peter Griffin teaching you how to work out. And let me show you how to do this.

How do you make Peter Griffin? Well, for that, we have to use Higgs field. I've been talking a lot about Higgs field in my videos, and that's because Higgs field really works.

I use Hixfield on my own videos, on my own Facebook channels. I even use Hixfield on the new Y. I create PNG photos.

I create motion graphics. I create all kinds of things for my content. It works.

It helps. It speeds up the process and makes it more efficient. So as I said, the first thing we need is Hixfield.

The second thing we need is Cloud Code. But by now, I assume you already have Cloud Code installed on your computer. If you don't, go ahead and get Cloud Code, create an account, and make sure you have Cloud Code on your computer.

The next thing you want to do is connect Higgs field to cloud code. For that, there's going to be a link down in the description of this video. It's the first link.

If you click on that link, it's going to send you to the Higgs field dashboard. Over here at the top where it says MCP, you click on MCP. Then over here where it says cloud, you click on cloud.

Then you just click connect to cloud. You're basically connecting Higgs field to cloud. Now, the reason why you want to do this is because instead of us prompting Higgs field to create the images, we're going to have Clot code do that for us because this way, Clot code can create much more detailed prompts than we do.

Not only that, but Clot can do it really fast. Within a minute, you have 20 prompts that are super detailed and can create the right character for you. It's going to create a character that has multiple poses.

So you're going to have a character that first stays like this and then like this and then like this. And then Clot code is going to add motion to it. which will make it look like a 2D animation similar to the Odds 1 Out channel.

But these images usually come with backgrounds. They come with white backgrounds or blue backgrounds. But Hicksfield has a tool called Background Remover.

And because we've connected Claude to Hicksfield, Claude can use that tool to remove the background from all your characters and all your assets super fast. So you end up with PNG images. You end up with images.

who don't have a background. You just have the character. And now because you have the character, Cloud Code can add the motion to it.

It can move it around, can change the frame so it looks like it's a 2D animation created by a human, which is really good. That's the workflow you want to use. And that's the workflow I use on some of the new freezes channels I've been creating lately.

And basically, once you've connected Higgs field to Cloud Code, just go to Cloud Code and type something like this, types of a prompt like this. I want to make a YouTube channel that teaches people how to work out and I want the main character to look like Peter Griffin from Family Guy, but a muscular version of him, wearing only pants and shoes.

Now, the first thing I want you to do is make the thumbnail and the title so I have the idea for the video. And you send this prompt to Cloud. Within 5 to 10 minutes, Cloud is going to create a bunch of thumbnails for you and a bunch of ideas for you with titles.

You pick the one you like and then you go ahead and send another message to Cloud asking it the following thing. create an eight minute script, and then create the animation using X field and generating the PNG images using X field. That will look like a 2D animation where we have Peter Griffin teaching us how to work out, you know, something simple like that.

You send this prompt and basically 20 minutes later, 30 minutes later, you're going to get an animation from cloud code, which looks like this. And let's react to it together.

Yeah, I'm impressed by this. It really looks as if an actual human created this animation frame by frame because you have the character which looks the same doing specific poses because As I said, Cloud Code was able to generate these images multiple times because we're using Higgs field and we can create multiple prompts.

It's just insane. Now, if you want to learn more about this and you found this fun to learn, there's going to be a link down in the description where you can join my private community. Inside my private community, I have a step -by -step course where I teach those things.

I teach how to do storytelling, how to hold retention, how to make people click, how to create your own niche. Not only that, but I answer to every single member. So basically, once you join...

you can send your channel and I will review it. I will help you improve the titles, the thumbnails. As you can see on the screen, I talk with my members.

I share feedback every single day. Every single day I wake up and for two hours in the morning, I just answer members and I share feedback. So if you want my personal help, there's going to be a link down in the description where you can join my private community.

The Hook

The bait, then the rug-pull.

The video opens on a real YouTube Studio dashboard: almost $2,000 in about three weeks and roughly 50 million views from a Minecraft animation channel. Everything after that is the case for why copying channels fails and what to do instead.

Frameworks

Named ideas worth stealing.

05:30model

Inspire, don't copy

Find a proven channel
Change one packaging element (character, palette, style)
Attract a different audience
Keep the topic the same

Keep what works about a proven channel's topic and structure, but change the character or look so the viewers differ. A different audience tells the algorithm the channel is original.

Steal forany faceless or saturated niche

10:40model