your ai model looks too PERFECT and every guy scrolls past her..
she reads as a render. no guy pictures her ever talking to him
a guy has no fantasy to step into with a girl who looks like that
so you build her so she doesn't read as a render
> generate 20 or more images of your ai model for one shot. throw most of them out, about 80%
> look at each one that's left for about 2 seconds and ask one thing. does she read as a real girl or as a render
> bin the renders even when nothing is wrong with them. a face with nothing wrong reads as generated
that 2 seconds is the same window a guy gives you on his feed
keep the one a guy would believe
comment "RENDER" and i'll send you the quick version of how i build my ai model
And here we go…GPT-6.2 Astra and GPT-6.2 Sol are yours to make more games with.
Oh, and they’re also super efficient and really good at making slides, building websites, and tackling the work you were doing before you started making games.
Rolling out today to ChatGPT Work and Codex for Plus, Pro, Enterprise, and Edu users.
https://t.co/tjQBDQWB2n
this is probably the lowest effort content that still does numbers
the less restricted cloud models like seedream keep messing up your ai model's face because they're just not as strong as the top ones
and you're sitting there tweaking it
cover her face with the phone
mirror selfie
done
nothing left to get wrong. and a girl holding her phone over her face in a mirror reads more real anyway
it's piss easy to prompt. anyone can do it
then the same still goes straight into wan or any image-to-video model
ask for almost no movement. no posing no zoom
now you've got a clip too
Introducing Claude Meetings: add Claude to your team’s video call and it builds docs, whiteboards, and other artifacts with you while you’re still talking.
Starting to roll out in beta on Team and Enterprise plans.
every time a new ai image or video model drops, stop stressing and rushing onto it...
a new release usually pushes down the price of the models already out
so the one you're running right now just got CHEAPER without you touching anything
> your ai content stays the same
> your costs drop
> your workflow doesn't break
for 90% of your ai content you don't need the frontier model
still worth a test. you lose nothing by trying the new one
carry on with what you're running. the frontier model is for the few generations that need it
I can test a dance reel for my AI girl without organising a shoot or performing on camera.
But a dance getting views isn't the same as her Fanvue making sales.
First, I check whether the opening holds attention and brings people to her Instagram profile.
Next, I make sure her other posts feel like the same character, aimed at older US guys.
Then, I track Fanvue subscriptions and extra photo/video sets bought by visitors.
I'd get that path working on Instagram before giving myself a second feed to fill on TikTok.
ONE image is all it takes to keep your ai model consistent every single time
the tool doesn't remember her from last time.
so every new picture or video it makes of her starts from scratch
one board of her goes in every time
> her face straight on close up
> her face at three-quarter
> full body from the front
> full body from the back
only the angle changes between the four.
the body matters as much as the face
you hand it that board each time so it has her whole look to pull from instead of inventing her
BEST way to keep her consistent
here's an example
ONE image is all it takes to keep your ai model consistent every single time
the tool doesn't remember her from last time.
so every new picture or video it makes of her starts from scratch
one board of her goes in every time
> her face straight on close up
> her face at three-quarter
> full body from the front
> full body from the back
only the angle changes between the four.
the body matters as much as the face
you hand it that board each time so it has her whole look to pull from instead of inventing her
BEST way to keep her consistent
here's an example
every ai video and image tool you're using right now is the cheapest it will ever be
uber pulled the same move. the first year or two the rides cost almost nothing, then the price went up
these companies are burning through investor cash to get you hooked
when it dries up the price jumps overnight... or the quality drops to match what you pay
cloud video gen is sitting in that exact window right now
whatever you've been planning to build or make for your ai model, do it NOW
so use and abuse video models while you can
once you've scaled and the prices start going up, move your work onto your own hardware
i had to check twice that this was my ai model
seedance 2.5 nailed every part that gives most ai clips away
the shaky front camera at arm's length, her looking up and away mid sentence, a guy answering from off camera, street light flaring the lens at night
it's starting to look like a real creator filmed it
my workflow one-shot it. Genuinely terrifying.
i had to check twice that this was my ai model
seedance 2.5 nailed every part that gives most ai clips away
the shaky front camera at arm's length, her looking up and away mid sentence, a guy answering from off camera, street light flaring the lens at night
it's starting to look like a real creator filmed it
my workflow one-shot it. Genuinely terrifying.
my airtable workflow is f*cking INSANE..
it's spitting out images and clips of my ai model that look like real life
i find a format that already went viral on another page and recast her into it. every idea comes back as 10 versions and i only post the best one
and her face comes from her saved reference images, so it's the same girl in every single post
comment "AIRTABLE" and i'll send you the full system to make these