Hang tight while we fetch the video data and transcripts. This only takes a moment.
Connecting to YouTube player…
Fetching transcript data…
We’ll display the transcript, summary, and all view options as soon as everything loads.
Next steps
Loading transcript tools…
EXACTLY How To Make VIRAL AI Videos Using VEO 3 | Dan Kieft | YouTubeToText
YouTube Transcript: EXACTLY How To Make VIRAL AI Videos Using VEO 3
Skip watching entire videos - get the full transcript, search for keywords, and copy with one click.
Share:
Video Transcript
Video Summary
Summary
Core Theme
This content provides a practical guide on how to create viral AI-generated videos, including memes, animal antics, and vlogs, by leveraging specific AI platforms and prompting techniques.
[Music]
These are the most viral AR reels that
And they are getting millions of views.
They make me smile every day that I see
them. And this is kind of like the AI
slop the internet was made for. So in
this video, I will show you exactly how
I will also show you how you can make
Honestly, I don't know who came up with
that, but it's super cool. Like, check
out this dude. He's getting insane
amount of views on his Instagram page
with these videos. I'll also share how
you can make these viral cats baking videos.
I will also explain how to make all of
these viral videos that are a bit older,
but still I will show you how to make
the Bigfoot one.
>> It's day three of solo camping in the Amazon.
Amazon.
Most of these are all made using a
single platform that you can use to make
both your images and your videos for.
There's one extra tool that we need that
I will show you throughout this video.
You can find the links to all of it in
the description down below, including
all of my prompts, everything I've used
if you sign up to my free newsletter.
We're starting off with this CCTV animal
videos that every time I see them, I've
been caught by surprise. So, let me
explain how you can make this. After
logging into Open Art, you want to go to
video. And for this video, we'll be
using text to video. Not image to video,
but text to video. You can do image to
video as well if you want to have a bit
more control. But if you're using V3,
then this works exactly. Now, you want
to enter in your prompt. So, if you sign
up to my newsletter, you will get access
to all of the prompts that I have. For
this first one, we're using this exact
prompt right here. Now, we're just going
to copy it, and we're going to copy it
into our open art. Then for the audio,
you want to turn it off. This will save
you a little bit of credits. And also,
we don't really need the audio for this
clip. If you want to use the audio, you
can do so, but it will be a bit more
expensive. You want to increase the
resolution to 1080p. And then you want
to do fast mode because that's also a
lot cheaper than normal mode. Now, you
click on create. And now your video is
As you can see, we don't have the part
where they put the thumbs up or the
middle finger yet because that's
unfortunately not really possible with
Google V3 Tree. You can do the thumbs
up, but I've tried multiple different
times, but it just turns out bad. Now,
to get the middle finger effect, you
need to go to the next step, which is
signing in to Pix. I will leave the link
in description down below. This is the
only platform right now that can
actually do this. Other than that,
everything can be done inside of
OpenArt, but for this, we're going to
Pixver. The next thing you want to do is
you want to go over to the template.
Then you want to scroll down and here
you have the dancing effect. But for
this we first want to use the middle
finger up. Then you want to upload your
open art video right here. So I'm going
to upload my footage right here. Then
you click open. And now all you got to
do is you have to click on create. And
then after generating you get this result.
If you want to add in the dance, then
you do that as follows. So, we replace
this effect with the dancing effect.
Then, you want to download this video
and you want to add this to your
project. So, I have it downloaded
already. Now, we have the middle finger
video added to this. And then, we just
add this at the end. And then, we hit
create. This might take you a few tries
because for me, it was glitching a
But honestly, this is quite good. And
these memes are a ton of fun. They add
in some automatic audio, but you can
just replace that with anything of your own.
Next up, we have this creepy AI account
that has some kind of like food
cannibalism going on. And honestly,
their views are quite insane. like this
one 13 million, this one 55 million, 36
million. So many people have brain rods.
Now to make something like that, we
first need to make an image. So you want
to go over to image on open art. And the
reason why you want to make an image is
because that will give you more control
over the outcome. So for the model, you
want to select flux context max. In my
opinion, this works best for making
these kind of like fantasy type
creatures. Then you want to add in your
prompt, which you can just copy my
prompt. You enter in the prompt right
here. I have it a hyperrealistic yet
cartoonist surrealism. A serving of a
golden French fries that have closed
eyes. Add in all the details about the
food that you want to add in there. Then
for the output size, in this case, I'm
going to do 16 by9 because this is a
YouTube video. If you want to do
Instagram, then you might want to switch
it over to 9x6 or 4x5. I must say 16x9
might be best because that will give you
more possibilities when generating the
video. So, we're going to go with 16x9.
And then you want to hit create.
That gave me this creepy dude. The next
step you want to do is you want to
upscale this because having upscaled it
will give you a better result. I already
have done that, but you can do basically
2x or 4x. And now that you have the
image upscaled, you want to make a video
out of this. So, you want to use the
image and then you want to go image to
video. This will bring you over to the
image tovideo generator. And now you can
add in your prompt. So for this, you
want to be using VO3 because VO3 is the
only one that can generate good sounding
audio with this. And also the motion and
everything combined is just great on V3.
Next up, we add in our prompt. So my
prompt to give you a bit of an idea is
hand with neatly manicured nails painted
with a soft pastel pink. And I even
included like the color code with that.
Enters frame from camera left gently
holding like a single golden French fry.
And basically we're feeding it the
French fry. It's it's kind of dark but
we're making food cannot loing videos
because that for some reason goes viral.
Like that's what the people want to see.
Now you add in all the details that you
have. You can just say this in talking
language as long as you're specific. So
the character eyes remain closed. The
mouth go opens naturally receiving the
French fries.
Okay. Then the character is tuned with a
joyful expression and steam is rising
softly from the fries. Then you add in a
few details like realistic motion, a
subtle head movements, and basically you
want to make sure you want to turn the
audio on, then you want to do 1080p, and
then you want to use fast, and then you
want to click on create. Now, that gave
I don't know about you, but I'm just
creeped out by this. But if you want to
do this, go ahead. You know how to do it
right now. You can do this with any type
of food. Let's for example do this with
a marshmallow. So here I got the prompt
for a marshmallow. First we of course
generate the image. Again I got my
prompt right here. Then we want to do it
widescreen. And then we have an image
Then we take that image and we put it in
the video tree and it gave us this result.
The last step you need to go through is
to add in a song. And the best songs you
can find on art list. So if you go over
to art list, you can find this song over
here. It has that creepy vibe. You can
just search for those songs. And then
you add this under your video. And that
The third meme that I want to show you
are these cat videos. Like they are
sprinkling some like flour on the cats
and then they're just massaging them
with a dough roller. It goes super
viral. Like people like it so much. This
dude got like close to a million likes
on that video, which is insane. And so
many people are thinking that it's real
So, let me show you how you make this.
For this, we first need to have an
image. So, with this one, we are using
uh not Flux Context, but we're using
Flux Deaf. And for Flux Deaf, you want
to enter in your prompt. I have a very,
very long prompt for this. Extremely
chubby orange tabby cat with a faint
cream stripes lying completely flat like
a loaf on a large wooden baking board.
evenly dusted with fine white flower.
Face is round with plush cheeks, short
muzzle, large perfect amber eyes looking
forward. Essentially, what I'm doing is
I'm describing the exact shot that I
want, like describing what type of view,
like a bird's eye view from directly
above. I describe what's going on in the
kitchen background. And I even add in
some texture and I say what lens it is
shot on. So, in this case, 85 mm lens
emulation. You can try this out with
Chetchip T. Copying my prompt, put it
into Chetchip T and ask it to recreate
these prompts with your idea. Then you
want to choose your desired output. And
I'm going to do 16 by9 again as this is
a YouTube video. Now we get a result
You then again want to upscale this
image. And then you know the drill.
We're going to put this inside the image
to video generator. For this one, once
again, we're using VO3, but you can also
try this out with different methods. But
I actually want to add in some sound
effects on this one. So, I'm using V3.
Then, we copy this prompt. And what you
essentially want to do is you want to
make sure you kind of like squish the
cat and then you sprinkle some flower on
it. And then you roll the pin over its
head. Essentially, describe what you
want to do in your prompt. So, you want
to enter in your prompt. Then, you hit
You can do this with different animals,
too, if you'd like. It isn't limited to
cats, but it is quite cute. Moving on,
we got AI ASMR videos, and these are
super satisfying. Honestly, this is just
better than a real person doing AI ASMR.
So, what you want to do with this, you
can do this yourself, and all you need
is V3 with that. The only thing you need
to come up with is a good prompt. Now,
let me show you how you can actually
make a good prompt for this. Basically,
what you want to ask it is to make an
ASMR cutting video. I normally use this
with my dictate mode, but to explain you
a little bit how that works is I just
say like make an ASMR cutting video with
glass strawberry on a wooden cutting
board. The video should be focused on
the satisfying ASMR glass sound. That is
very important. There should be a unique
distinguishable feature. So, the woman's
hand holding the strawberry has like
pink nail polish. Now, this will make it
more unique. If you want to have your
own page around multiple different AI
ASMR videos, for example, then you can
say, "Give me a hex code for the color
to maintain consistency." Also, the
knife is a polished stainless steel
kitchen knife, and the handle has a
distinguishable feature, which is an
emerald gemstone handle. This makes your
prompt more unique than others. You can
ask it to give the hex codes so you can
maintain consistency. Then, you hit
enter. Now, it will help you structure a
good prompt.
We can now copy this prompt. And now we
go over to open art and we're going to
use text to video. And here you want to
make sure you use the video tree with
audio on. Then you click create and that
You can do the exact same thing with a
glass pancake for example. So take a
Next up, this gummy bear is also super satisfying.
And you can find all of these prompts
right here. Another thing that's still
going viral and will help you a lot with
structuring your prompt is Jason
prompting. Now, these IKEA unboxing
videos have been performing really well
on the internet. People might already be
over it, but still learning how to JSON
prompt can be super useful for making
cool videos. It is essentially a great
structure of telling the AI exactly what
you want to have happen. Now, I've made
this custom chatd prompt that will help
you make a good JSON prompt just by
typing in what you want. For example,
let's say you want one for a high-end
perfume. Now, it will start writing the
entire prompt for you. Ideally, you want
to be more specific. You want to add in
all the details that you want, even if
it is something custom like a certain
object that you want to have in there.
Make sure you describe it as clearly as
possible. Then you copy your prompt.
Then you put it inside of V3 text to
video. And then out comes your video.
Lastly, the Bigfoot/ Yeti vlogs have
been going viral for quite a bit of time
now. And let me just show you how this
works. In my opinion, it's a bit
outdated right now, but still people
want to know how that works. So what you
want to do here, you want to create your
prompts first. I am a big fan of using
Chachi to write your prompts because it
can add in more details than you can
imagine yourself. The other thing that
you can do when writing your prompts is
adding in hex codes. Rookie made a good
video on this. So shout out to Rookie
for coming up with this idea. And what
you essentially do is you copy the code
that you want to have something like the
fur of Bigfoot or some kind of object.
You add in that exact color code of what
you want it to look like. Then the AI
can understand, oh, he wants it to look
like this. So, no more guessing around,
getting precise prompts to get exactly
what you want. Basically, what I do is I
just dictate it like this. I want to
make a POV vlog style video of Bigfoot
holding a black GoPro Hero8 and he's
essentially vlogging. Could you add in
the hex code for this and also describe
what it looks like, where he is, in what
type of location he is? For example, it
could be in the woods, could be in the
forest, could be anywhere that you can
imagine. Then we hit send.
So now we have this draft. I don't like
it that much. So we want to change it a
little bit. The things I don't like is
it needs to start with POV. So let's
just instruct it. Start the prompt with
POV vlog style video of a hairy Bigfoot
with the color codes of his fur. Then
also include that he is holding a GoPro
Hero8 that is black with the hex code
and make sure he is like let's do let's
put him on a on a beach at sunset. Now
make him say yo bigfooters this is day
number three and I am still surviving on
this island. Those monkeys over there
have not been nice to me so I'm going to
beat their ass today. Next up you want
to hit send. And now out comes our
prompt. So that's essentially how you
can kind of like learn how to prompt
with chatbt. Let's copy this. And now we
put this over into open art. And we're
using text to video and v3 audio on.
Same settings fast. And then you hit
send. The one thing you want to watch
out for with Google V3 is that it is way
too censored. So I get like a content
policy violation right here. The way to
solve that is to probably avoid this
wording right here, which is going to
beat their ass. So fortunately, Google
has been playing the nice game and you
cannot like say anything weird in it. So
let me just change this to I'm going to
throw rocks at them today. Now we try it
again and let's see if this get
approved. So my text here is way too
long. Your limit is around 8 seconds. So
make sure you fit in the exact text that
you can fit within 8 seconds. Here's
another video that I made. You can find
the exact prompt in the prompt document
that I've attached. Now take a look. Hey
Bigfooters, it's day three of solo
camping in the Amazon.
If you want to try all of this yourself,
then I would recommend signing up to
Open Art. It is the all-in-one platform
where you can do multiple different
things from generating images to videos
to using all the different AI video
platforms. And also, if you sign up to
my free newsletter, then you will get
access to all the prompts that I used in
this video. So you can try generating
this yourself, use my custom prompt bots
and everything you need to make cool
videos. If you're still conflicted to
which model you should use what for,
then click the video that's on the
screen right now. And in that video, I
will dive into how to use each and every
AI video generator that's on the market right
Click on any text or timestamp to jump to that moment in the video
Share:
Most transcripts ready in under 5 seconds
One-Click Copy125+ LanguagesSearch ContentJump to Timestamps
Paste YouTube URL
Enter any YouTube video link to get the full transcript
Transcript Extraction Form
Most transcripts ready in under 5 seconds
Get Our Chrome Extension
Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.
