Faceless YouTube, made by hand and by model

What a faceless channel is, how one video is made from start to finish without a camera, and what YouTube asks of AI-made videos.

The topic

A faceless video is one where nobody stands in front of a camera. The story is carried by a narrator's voice, by pictures and by motion. Some of the most watched explainers, history channels and story channels work this way.

We run our own faceless channels through Cinetana: pregnancy stories, home repair, survival and campground guides among them. They are small. The largest had 137 subscribers on 13 October 2026. What we can show is how the work is done, not that it grows a channel, and that is what this topic covers.

In the studio

Step 01Pick one promiseEach channel has one audience and one promise, written into its channel settings in the studio so every script starts from the same brief.

Step 02Write for the earScripts are drafted by a Claude model, then rewritten line by line for speech. A person reads every fact and every title before anything is made.

Step 03Picture each lineEvery sentence gets a still made with Google's Gemini image model, then the strongest stills are moved with Kling, a third-party video model.

Step 04Label and uploadThe cut gets captions, a thumbnail and chapters. AI-made realistic scenes are disclosed in YouTube's upload settings, every time.

The guides

01
Faceless video AI: how one video is made, end to endThe whole chain, idea to upload, on one of our own videos.
In writing
02
Faceless YouTube channel: what it is and how it worksFormats that work without a face, and what YouTube expects.
In writing
03
Faceless YouTube channels: examples by formatLed by our own channels, with real, dated numbers.
In writing
04
How to start a faceless YouTube channel: the first weekNiche, channel setup, first three videos, AI disclosure.
In writing
05
YouTube automation tools: the jobs, not the brandsResearch, script, visuals, edit, thumbnail, upload.
In writing
06
How to do YouTube automation, and what a person still checksWhat we automate, and the rules on repetitive content.
In writing
07
YouTube automation meaning, in plain wordsWhat it is, and what it is not: reuploads and spam.
In writing
08
YouTube automation business: an honest look at the costsWhat a video costs us to make, from our own cost meter.
In writing

Rules we keep

No real person's face or voice without their written consent.
No reuploads of other people's videos, and no music we do not hold a licence for.
No promise of views, subscribers or income. Our numbers are shown dated, as they are.
Realistic AI scenes are disclosed to viewers, following YouTube's own rule.

Sources