Frequently Asked Questions

Commissioning an AI short film

How producing a cinematic film without live-action photography actually works, how long it takes, what it costs and who keeps the rights — clear answers to the most common questions.


Can you make a short film without any live-action camera footage?

Yes. An AI-produced short film requires no camera, set, cast or location — every frame is generated.

What changes is the tooling, not the directing. A scene still only lands emotionally if the script, shot order, lighting logic, editorial rhythm and sound design are right — the entire discipline of classical filmmaking.

FLAWLESS (a 4K documentary of roughly 17 minutes) and the Bushido concept brand campaign were both produced this way, with zero photography.

Can I turn a personal story — a relationship, a family memory — into an AI short film?

Yes, and it is one of the most requested types of work.

The process starts with a conversation about the real moments: which scenes are unforgettable, which detail makes this story different from any other. Photos, messages and voice notes can be used as reference. Those are then structured into a script and produced scene by scene.

The goal is not literal documentation but making the audience feel what you felt. Birthdays, proposals, anniversaries and memorial films all fall into this category.

How long does a 10-minute AI short film take to produce?

With a locked script, a 10-minute film typically takes 3 to 6 weeks.

Roughly: 1 week for script and storyboard, 1–3 weeks for image generation and scene consistency, 1–2 weeks for sound design, music, editing and color.

If the script is still in development, add a 1–2 week development round up front. Rush delivery is possible but usually means sacrificing scene variety.

How much does an AI short film cost?

Three things drive the cost: runtime, number of distinct scenes, and how much character consistency is required. A 3-minute film in one location with two characters is a very different job from a 10-minute film across ten locations.

Compared to conventional production, the biggest saving is that set, crew, equipment, cast and location costs disappear entirely.

Share the intended runtime and how many locations and characters are involved, and you get a price range the same day.

Do characters stay consistent between scenes? I heard AI struggles with that.

This is the technical issue that separates amateur work from professional work. Raw models produce a slightly different face every time.

The fix is to build a locked reference set for each character and derive every scene from it: fixed character references, wardrobe and lighting continuity rules, shot-level review, and post-production correction where needed.

In the Bushido campaign the synthetic actors held the same identity across three separate films.

I don't have a finished script yet. Can we still work together?

Yes. Most projects start with an idea or a feeling rather than a finished script.

Script development is built into the process: we talk the story down to its spine, turn it into a scene list, and only then move into visual production.

This is where working with someone trained in cinema matters — operating the tools and building a story are two different skills.

Which AI tools do you use?

Tool selection changes per project, because each model is strong at something different.

The stack combines image and video generation, footage enhancement (Magnific), voice cloning and multilingual dubbing (ElevenLabs), music generation (Suno — I am an official Ambassador) and conventional post-production tools.

What matters is not which tool is used, but knowing which task belongs to which tool — and unifying the output into a single visual language.

Can the film be voiced and lip-synced in other languages?

Yes. Multilingual voice-over and lip-sync are part of the standard service, including Turkish and English. The same film can be delivered in several languages keeping the same vocal character.

For the FLAWLESS documentary a complete Italian dub was produced in my own cloned voice; Turkish subtitles are live, with German, Spanish and French subtitles on the way.

For international campaigns, this is how a single production is adapted to multiple markets.

Who owns the copyright and commercial rights to the film?

The delivered final film and its commercial usage rights sit with the client, with the scope written into the contract.

Music is either licensed (from sources such as Epidemic Sound) or composed for the project, so there is no clearance risk from pre-existing tracks.

Personal material shared as reference — photos, audio, messages — remains yours and is never used anywhere else without permission.

What does the process look like, and how many revisions are included?

Four stages: (1) a short online call and brief, (2) sign-off on script and scene list, (3) a visual direction test — the film's look approved on a few sample frames, (4) full production and edit.

The approval gates sit deliberately early — the most expensive mistake is discovering the film isn't what you imagined after it's finished.

Two rounds of revision at the edit stage are included as standard.

I'm based outside Türkiye. Can we work remotely?

Yes — a significant share of the work is remote. Because the production model needs no set or location, geography isn't a constraint.

I work in Turkish and English; briefs and reviews happen online, and delivery is digital.

How do I get started?

Email eray@eraybalat.com with a few sentences about your story. Include the rough runtime and how many locations and characters are involved, and you'll get a timeline and budget range in the first reply. From there, a 20–30 minute online call locks down the project.

eray@eraybalat.com

Looking for brand films, product films or UGC ads instead? Those have their own page with services, timelines and pricing: Hire an AI filmmaker →