How a stickman video gets made

It learns the topic before it writes a word

Most AI video tools write the script straight from a one-line request, from whatever the AI happens to remember. YourFacelessFactory researches first.

• Live web search. Before any writing starts, the topic is researched with live web searches: up to 25 for a long video, fewer for a Short. The goal is real facts, real numbers and fresh examples, not the same five talking points every other video on the topic repeats.
• Reworking a YouTube video. The video's transcript is read to understand its subject and angle. It is never used as the script. About half of its points are kept in completely new words, the other half are replaced with newly researched facts, and the two are woven together so the result stands on its own.
• Starting from an idea. A one-line idea is researched from scratch. Paste a longer brief (notes, an outline, even a draft) and it's treated as your content: the facts in it are checked against search, anything wrong is corrected, and gaps are filled.

Research makes a script far more accurate than writing from memory. No AI is perfect, though, so skim the important claims before you publish.

Written like a real channel, and written for the ear

The script is written with the same structure successful explainer channels use, as an entirely original piece of writing.

• A hook that earns the click. The first seconds open with what happened, not the topic, and save how it happened for later.
• Mini-stories that keep people watching. The middle is built from three to six short stories, each with a setup, some tension and a payoff, woven into one story instead of a list of facts.
• Comparisons you can picture. Big numbers are turned into something physical and relatable, and earlier moments are called back later.
• Made to be spoken. Numbers, ranges and units are written the way a narrator says them, units are converted to metric, and abbreviations are spelled out, so the voice never stumbles over "92x" or "10-15".
• Every line is also a picture. Each scene becomes one picture of a planned kind: a story scene with the narrator, a map, a labeled diagram, a comparison, a big number, or a bold text card for a twist.
• Pacing you choose. Four pacing levels, from a new picture every one to three seconds to a calm one every eight. The script is written to that pace, then sentences that pack in several ideas or run long are split into separate scenes (or short scenes are joined, at the calmer levels and at the fastest). Every split is verified word for word, so no narration is ever lost or changed.

A natural-sounding voice, recorded in long, flowing pieces

• Your choice of voice. Narration is recorded with ElevenLabs, in the voice you pick and preview in the template builder.
• Long recordings, not clip after clip. Scenes are recorded together in long, unbroken pieces and then cut at the exact word. The delivery flows naturally across scene changes instead of restarting its tone every few seconds.
• Everything is timed to the words. The exact time of every spoken word decides how long each scene lasts, when each part of a list or a diagram appears, and where every caption starts and ends.
• The length you asked for. After recording, the real duration is measured. If it falls outside your target, the script is rewritten at a corrected pace, up to four times.

Every line gets a planned shot, not just a picture

• Planned with the whole story in view. A planning step goes through the whole script in large batches and picks one picture for every scene from a library of kinds: a story scene, an establishing shot, a close-up, a map, a labeled diagram, a cross-section, a comparison, a big number, a text card and more. It fills in what each picture shows and the few words worth writing into it, so a figure of speech gets drawn as what it means, not literally.
• Shots like a film. Wide shots, full shots, waist-up shots and close-ups take turns. When a moment goes on in the same place, the next picture shows the same scene from a new camera angle (closer, from the side, from behind) instead of a near-copy.
• Words drawn into the picture. Names, labels, dates and numbers are drawn into the picture itself, in the art style, only where they earn their place, and every word is read back and checked.
• A camera with intent. The camera moves only on story pictures without words: a slow glide across a place being described, or a slow push-in. Now and then it zooms in quickly on the exact thing being named, right as it's said, for emphasis; an AI finds that thing in the picture first. Pictures with words, numbers or a diagram always hold perfectly still, and one switch turns all camera motion off.
• Pictures that explain. Places get a map. The inside of something gets a cross-section or a labeled diagram. "Wrong vs. right" gets one picture with an X and a check mark, two things side by side get a "versus" picture, and a list or a diagram builds up part by part as the narrator names it.

One art style. One cast. Every scene.

• Drawn from scratch, line by line. Every scene is illustrated for the exact line being spoken. There's no stock footage and no clip-art library.
• A strict set of style rules. Pick any of the thirteen styles on any plan (Doodle, Creepy Comic, Minimal Doodle, Classic, Ink Noir, Cinematic, Gritty, Minimal, Highlight, Brushpen, Cozy, Vivid or Glow), each drawn by its own rules. The style shapes the world: its lines, colours, light and places. Your characters are drawn exactly as they are in every style, never restyled or tinted by the scene's light. Every single picture the AI is asked to draw comes with the rules of the style you picked.
• Your narrator always looks the same. The main character is drawn from its reference pictures in every scene (the stickman's show every angle): the stickman, one of our 24 other characters, or your own, made in the character chat or from a picture of yours.
• Every important character always looks the same. Before their first scene, each important character gets their own reference portrait. Real people from history are researched for their clothing, headwear and signature props. Recurring groups get a group design sheet, and recurring objects and animals keep the same look after their first appearance.
• Lists that build up. When the narrator lists things or names the parts of a diagram, the picture is drawn first and then edited one part at a time, so it stays the same picture as it grows.

It checks its own work

Generating an image is the easy part. Knowing it's the right image is the hard part, so every image is checked by a different AI than the one that drew it.

• The right subject. Does the picture show the specific thing that was asked for, or a vague, generic stand-in?
• The right character. Does a recurring character match their reference: head, hair, build, clothing, and even the number of arms and legs?
• The right words. Is every word drawn into the picture spelled exactly right, with nothing extra written anywhere?
• Drawn again when it's wrong. A picture that fails the check is drawn again with what was wrong spelled out, and a wrong word is fixed. A picture that's still off after that is named on your video's page, and Edit scenes can redraw it.
• Stops before it wastes your tokens. If the picture service breaks down (five pictures in a row fail), it stops instead of spending your tokens on a video that won't come out right. And a video that fails on our end is never charged.

Finished, not almost finished

• Built at full size. Pictures, camera moves, captions, music and narration are put together into one video at the platform's own recommended size: 1920×1080 landscape, 1080×1920 vertical, 1080×1080 square or 1080×1350 portrait.
• Captions timed to the voice. Optional captions built into the video follow the narration word by word: nine animated styles to start from (a box that slides from word to word, a highlighted spoken word, words that pop in and more), and every detail adjustable.
• Thumbnails built to be different. Optionally, several thumbnail concepts are planned together so they don't all say the same thing, drawn at a higher resolution, and checked for correctly spelled title text. Upload a reference and they'll follow its look.
• Ready for YouTube. A title, a description, five hashtags and chapter timestamps are written from the video's own script, with chapters placed where each scene really starts.
• Download or autopilot. Download one zip with the video, title, description, thumbnails and narration audio, or link the template to a schedule. Scheduled videos are made up to a day ahead and posted to your channel right on time.
• Fix any scene afterwards. For 30 days after a video is made, Edit scenes lets you pick any scene, type what to change or have a brand-new picture of any kind made in its place, and Update video puts the new pictures in. Only the changed scenes are rendered again; the narration, captions and music stay exactly as they were.