Soku AI
All tools

AI Human Generator Cast a Person Without Booking One

Describe a person and get back a photoreal still of them.

Photoreal PeopleNo Shoot

Three deliberate steps

01

Describe who you need

Age range, wardrobe, expression, setting, and how tightly the shot is framed. "Woman in her fifties in a garden centre apron, mid-laugh, waist-up" beats "happy woman" every time.

02

Generate the still

GPT Image 2 renders the person at the aspect ratio you pick. Run the prompt again with one detail changed to get a second casting option rather than a second pose.

03

Ship it as creative

Download the still, or send it into Creatives to add the offer, the logo and the placement crops the buy actually needs.

Use cases

Who reaches for this, and when

Cast a demographic you cannot find in stock

Stock libraries are thin everywhere outside the 25-to-35 bracket, and thinner still once you need a specific occupation, region or body type. Describing the person you actually sell to is faster than filtering twelve thousand results for someone who nearly fits.

Fill the people-shaped hole in a product-only ad set

Most e-commerce brands have flawless packshots and no human imagery at all, which caps how far a prospecting set can be tested. A generated presenter gives the set a face to test against the pack shot without commissioning a lifestyle shoot.

Localise a creative concept per market

The same layout, the same offer, a person who reads as local in each market. Re-running one prompt per region is the cheap version of the shoot you were never going to fund in six countries.

Questions before you run it

Can I generate a specific real person?
No, and you should not want to. The model composes a person from your description rather than reproducing a named individual, and prompting for a real person — a celebrity, a competitor’s founder, your own customer — is both unreliable here and a fast route to a rejected ad and a publicity-rights complaint. Describe the type of person you need instead of naming one.
Do I need a model release for a generated person?
There is no individual to release, because no real person was photographed. What does still apply is platform disclosure: Meta, Google and TikTok all have rules on labelling AI-generated imagery in ads, and those rules track the ad, not the model that made it. Treat the release question as answered and the disclosure question as open.
Will the same person come back on the next run?
Not reliably. Each generation is independent, so a second run of the same prompt returns a plausible person, not the same one. If a campaign needs one consistent face across several frames, generate the hero still once and take that image into an editing tool that works from a reference, rather than expecting the generator to remember.
How well does it handle hands and text?
Hands are much better than they were and still the first thing to check at full size — especially where a hand meets a product. Any text in the frame is worse: treat rendered signage, packaging copy and labels as decoration to be replaced, not as copy you can ship.
Is the AI human generator free to use?
Sign up on this page and you get free credits to run it here — there is nothing to download. After that each generation is credit-metered, and the exact cost is shown before you submit, so a casting session cannot quietly run away with a budget.