I Tried To Build A $1M AI App

Startup Sphere
22 followers
19 Video Views·Sep 8, 2025
Starting completely from scratch we try and make $1M with an AI app.
Im at a crossroads with this channel. Do I keep documenting the process and make episodes every 2 weeks until we make $1M a month. OR still do videos on other topics. I would love your guys input
Five million in revenue.
Six million.
ARR.
A million dollars a month in revenue.
One million dollars every month.
And how old are you?
I'm 17.
This is going to be one of the craziest episodes I've ever done.
All of these guys are making millions of dollars off AI apps.
And Luke and I wanted to see if we could do the same thing.
So we're going to try and make a million dollars starting completely from scratch
to see how hard it really is.
Over the next three months, we're going to figure out an idea, get it built,
make a marketing plan, and then launch it.
How hard can it be, guys?
How hard can it be?
Over the next week, Luke and I spent every day brainstorming what we should build.
Between Luke and I, we've gotten billions of views across social media,
and every single dollar we've ever made has come from either our own content
or distributing other people's content.
We literally run the same playbook every single time.
Find what's already working, change it slightly, and make money.
It's pretty simple.
We call it the Zoinkin Twist.
It's how we've won on Snapchat, YouTube Shorts, Fortnite, YouTube Long Form.
We use it on pretty much everything.
So when we started brainstorming what we wanted to build,
we wanted to put that idea into some sort of software product
to see if we could make it easier for others to do the same thing.
And this is what we're thinking.
Basically, a short form video editor like CapCut
that has built in templates for the most popular formats.
Insert the Crayo logo.
Templates for the proven viral formats.
And I'm not talking about Reddit stories and text messages.
I mean like the real metas and foundational short form style,
like voiceover videos or the edited clips,
the stuff that's a little bit harder to do, but is way more consistent.
And then on top of that,
we can have AI tools for scripting voiceovers and captions.
We also came up with this feature to remix other people's videos.
Basically, you insert a link to another popular short
and we'll help you remix that video by slightly changing the words in the script,
keeping the good components of why it worked
and then allowing you to slightly tweak it or change it to make it your own.
And let's be real, guys.
Copying is what everybody does.
You don't need to copy a word for word,
but you can change something slightly about a working concept
and have much better chances of succeeding.
This is what everybody does all the time, but it is controversial.
So **** it.
And this is one of those things where if we don't make a good product,
we really don't make money.
I know this kind of maybe looks like,
oh, this guy's just trying to make a million dollars.
We don't want to make this like some gimmick.
If we wanted to do that, we would honestly just sell a high ticket course.
And maybe that's the play.
Honestly, maybe we should be selling a course.
But ****. Crypto coin?
I don't know. Let's see if this works.
So now that we've got the idea locked in, now we just need to build it.
The only problem is we're trying to build a complex software product
and either Luke and I know how to code.
So right now, the plan is just to reach out to every kid
we can find at Stanford who knows computer science
and try and pitch them on working on this project.
Is this retarded?
Maybe. This plan sounded way smarter in my head before I started recording.
Over the next few days, we had around 15 calls with Stanford devs.
Eventually, we found a duo who were down to take on the project.
They weren't looking for full time jobs,
but they liked the idea and were down to build a basic version for a flat fee.
They said the build itself wasn't going to take long,
which was honestly a nice surprise.
So we're just paying them right now to build the first version,
just enough for us to start testing with real people.
And then once that's done, we can bring someone on more full time.
So now that our devs are in motion, cooking up the early stages of this product,
Luke and I think it's the best use of our time to heat up our personal brand channels.
We think if we just provide as much value as possible and make entertaining videos
that when the time comes, everything else will work out.
I don't know if that's dumb or not, but we just kind of feel that way.
That process ended up taking way longer than I expected,
but the crazy part is the videos are actually doing well way sooner than I expected.
Personal channel. Let's check it out, people.
Look at that growth. Look at that growth, man.
A cave views on first day was lit.
22 days, 62 days. 62 times more views.
But little did we know, things were not going as good as we thought.
The next day, I just woke up and our devs said that they backed out
because of Stanford finals. What the **** is going on?
Like they didn't know finals was coming up and all of a sudden they can't do it.
OK, after losing a month of progress, we are not going to go for someone who is part time
in college and just get someone who is fully dedicated.
We hit Twitter, Discord and even Twitch chats,
literally messaging in the software live stream, trying to find someone who could build this.
But what we realized is the best developers aren't looking for jobs.
They're already working.
So the best thing we can do is just spam social media and hope to find someone good
who's in between projects.
So we cooked up the most compelling tweet we can think of to excite a developer.
And now we're just waiting.
OK, guys, so the tweet went out two days ago and it got a whopping 10 likes.
However, we did get one guy to reach out named Kush,
who's built creator software before and even started his own YouTube channel.
And he says he's down to take on the project.
So first things first, we had him do a little test to make sure he could
actually do the developer ability he said he could do.
Then he ghosted until he came back the next day.
And not only did he do that little test we said to do,
but he completed an entire section of the software.
So at this moment in time, we knew it was time for marriage.
I got down on my knees.
We gave him a fat chunk of the company and he flew out to L.A. to do this full time.
He's actually in the room right over there.
Oh, my gosh.
I'm just excited to actually start building something.
Things are really starting to come together now with this software build.
And we're seeing these features finally come to life.
On top of that, our channels are growing and on track to do really well.
So we are actually so back.
All right.
So it's been a lot of times since our last check in,
and we got way deeper into the build.
We've also talked to a lot of different people and business associates,
and we've realized a lot of flaws with this idea.
First thing is that our competitors are loaded.
Literally every one of our big competitors either raise VC money or is already giant
and is quite literally in a laboratory trying to figure out the future of A.I. editing.
The other thing is paid ads are probably not going to work.
Our product is going to be like $19 and everybody else trying to reach
our same audience is selling a $300 course.
So it's going to be way too expensive for us to pay to reach that same customers.
Even if we paid $18, these people would probably pay $40 to acquire the same customer.
And our ---- costs $19.
So paid ads are probably not going to work out for us that well.
Also, this is taking way longer than we expected,
probably because we're stupid as **** and thought building and editing software
was going to take way less time.
We're a few months in and we haven't even like seeded this out to customers at all,
which is usually not what you want to do.
But instead, we went three months down this product development journey
without even knowing if customers want this.
So yeah, there's just a lot of things.
Okay.
There's just a lot of things.
With all of this realization that doubts honestly started creeping in a little bit.
But at the end of the day,
**** it, we bought.
If not me, then who?
Used to be in my bio for five years.
We decided, let's just try and use the product on a fresh YouTube channel
and see what happens.
Meet Fam, a friend of ours who currently has zero subscribers
and is making zero dollars a month
and wants to do everything in his power to blow up a YouTube channel.
And thank God, agreed to use our editing software.
All right.
So what channel are you going to start, bro?
I don't know.
I'm thinking like Kaizenet.
I show speed.
You mean like the slam dunk format where it's like,
or like, what do you mean with the Kaizenet?
Yeah, I think slam dunk.
But is that like, is that still good on YouTube?
They say it's saturated, bro.
It's oversaturated.
They do say that.
And I feel like a lot of people are making Kaizenet clips and stuff,
but on the go, right?
On the go, right?
Come on.
It's a big, it's a big niche.
Yeah.
All right.
So if you're going to do that kind of channel on VueMax,
just click the B roll with voiceover.
It's basically the templatized version of that entire workflow.
Basic editor, bro.
If you know how to edit on anything else,
this would be like 10 times easier.
So I'm going to let you chef and do your thing.
And if you don't have 10K subs in a month,
then somebody's going to get spanked.
Me?
It could be me.
Oh, it could be you.
I hope you don't hit it, bro.
I won't.
I only hit 9,999 subs, bro.
No.
Over the next few weeks,
Luke and fam are masterminding how they could cook up videos.
And they decided to copy a format Luke had done on an old video
that got 90 million views.
It was a video pointing out primitive survival videos were fake,
but instead they replaced it with Ashton Hall videos were fake.
And the video is going ballistic.
All right.
Just got off a call with fam.
I'm in the lamb truck, by the way.
Honey putt.
Put a Lambo logo right there.
Basically, he's had a few flops since that banger.
And I keep telling him if he sticks with it,
he's going to have a certified hood classic soon.
But he's kind of mentally in a rut at this very moment.
So let's see how this goes.
And with everything going to ---- at the moment,
nobody could have expected what was about to happen.
He's averaging 40 for 18, 19 games.
He's dropped 35, 29, 50, 32, 47, 35, 35, 39, 41, 45, 41.
43, 44, 38, 32, 42, 43, 38, 57, 58.
Do you understand what you're witnessing?
This was our proof that our tool was actually valuable
and is motivating us to make way more templates.
Now with a product in a case study,
we just need to figure out how we're going to market this.
A couple of months ago, Luke had this tweet that went viral.
It got like over 10 million views and was basically him
describing how he started his slam dunk channel from scratch.
And if you look in the replies, everybody is hating.
Let me just read a few.
So recycle other people's popular content.
Steal the content for engagement and hack the algorithm.
His channel is literally regurgitated slop.
He is so young, rich, and handsome.
And the answer to all that is yes,
he is quite literally trend jacking other people's big content
and just telling stories about it.
This is what 70% of all content online is.
And this is probably the easiest way to make money
without putting your face on a video.
The only reason to actually be mad is if you think it works.
We think we can create the biggest rage bait campaign of all time
because every single person that tweets
is admitting that this ---- works.
And we're going to put it all on one launch day and see how it goes.
Luke's video with fam about the channel blowing up
will be the first launch case study.
We'll probably also pay some influencers and big pages
to push the marketing plan as well.
And finally, the baseline of the software is done.
And Loki took six months longer than we expected
because we're retarded, but we're about to push it live.
Three, two, bucket one.
Hit that bitch.
There we go.
Deployment created successfully.
We've made it, bro.
We've made it.
Hold on, I'm going to do a flip.
You know, it's all in God's plan, man.
Wait, can I buy this ---- right now?
Let's go.
Our first real buyer, our first real customer, guys.
