SL-23 / Surprise Without the Pose

The Piece Finally Fits

Three failed tries, then it clicks. The face is the last thing to find out.

The brief

After three failed attempts, the piece finally clicks into place. The hands stop before the face catches up, and the exhale arrives a beat late.

Why it's hard

The model wants the catalogue reaction — the wide-eyed "Wow" delivered by the whole body at once. Real bodies report news out of order.

The craft key

Stagger the reaction in three beats: hands stop, then breath, then face.

Skill Tests

04 / 28
Animal anatomy Human anatomy Biomechanics Physics & mass Water simulation Crowd dynamics Multi-shot continuity Face consistency Camera movement Unbroken take Motivated lighting Cutting rhythm Hands & fingers Hair & fabric Reflection & refraction Scale & depth Impossible geometry Period accuracy Music sync Aging & time Natural asynchrony Eye-lines & gaze Particle behaviour Motivated light source Consequence & persistence Time manipulation Graphic-layer discipline Colour as storytelling

Every challenge is scored against the same vocabulary, so you can compare them. Lit cells are what this one tests.

Starter prompt

Built with the CEMI MediaMax system for Google Veo 3.1. Prompts stay in English — that is the language these models take instructions in. Treat it as a starting point and start varying it.

the-piece-finally-fits
A woman in her forties works a hand-cut wooden puzzle at a scratched pine bench. Three real failures first: she rotates the piece, presses, it lifts and rocks; on the fourth try it drops flush with a dry click. Her hands stop dead and stay flat on the wood for a full beat, then the shoulders drop and the exhale arrives, and only after that does the face change. No hands to the head, no gasp, no held pose. Observational register of Hirokazu Kore-eda's Shoplifters: camera at bench height, handheld with an operator's breathing drift and one small reframe that arrives a beat late, as if he was watching the hands and missed the cue. Dirty foreground: a chipped enamel mug and loose sawdust across the near edge, thrown out of focus. Vintage 50mm Zeiss Superspeed at f/2, soft halation off flat overcast window light, no fill. Audio: the click, the wood, the breath.

Derived from the CEMI MediaMax production canon — cemi.media

Style reference: Hirokazu Kore-eda (Shoplifters)

Why it's built this way

The anti-AI moves here are staggered reaction timing and hands-not-eyes as the emotional carrier: the body finishes the task before the face is allowed to comment. Kore-eda is the reference because his camera treats domestic hands as the main event and never rushes a beat. Vary the object (a bike chain, a tent pole, a stubborn zip) and the failure count, but keep the order hands, breath, face.

Watch for

The model will fire all three reactions on the same frame — hands, breath and a broad smile simultaneously — which collapses the stagger and reads as a stock "Wow".

Ran a challenge? Show us what you got.

Share your attempt — the triumphs and the glorious failures. The most instructive entries get featured on the board.