Your free prompt is here
Copy it, paste it into your video model, and watch what happens. One thing first though โ scroll down before you hit generate.
Copy. Paste. Create.
Cinematic [X]-second continuous shot, 35mm film grain, shallow depth of field, warm natural color grade, photorealistic. The camera is positioned at eye level directly in front of the subject, front-facing POV, as they walk straight toward it down an upscale city sidewalk in [describe outfit/styling here]. They speak naturally and fluidly, in a quick, casual conversational rhythm with no pause, in English: '[your line here].' A stranger rushes into frame from the side and collides with their shoulder. THE COLLISION AND SEPARATION MUST BE CLEARLY VISIBLE AND UNHURRIED: the moment of impact eases into a brief, smooth slow-motion beat โ their shoulder and torso rock back from the hit, and in full, unobstructed view of camera, each held object leaves their body on its own clean, natural arc, drifting outward toward the camera in graceful, physically believable trajectories, nothing sudden or jump-cut. As each object reaches the peak of its arc, time freezes completely and everything hangs suspended in mid-air, their body caught mid-stumble. The camera whip-zooms briefly between the frozen objects one by one. The freeze ends abruptly, the camera zooms back out, and time resumes โ all objects fall to the pavement in the exact same instant the subject stumbles and falls to the ground, all landing together in one simultaneous natural fall. No visible logos or brand names.
Your reference image decides everything
The prompt is the easy part. What actually decides whether this comes out cinematic or chaotic is the reference image you feed the model. Lock that in first โ then the video falls into place. That’s the first thing I teach inside the Academy.
Skip straight to the video with a rushed reference image โ inconsistent face, broken hands, messy motion
Lock in a clean, consistent reference image first โ sharp face, clean motion, a video that actually looks real
A few months ago I didn’t know how to make a single AI video. Today I have a community of more than 60,000 people, and my income comes from creating AI videos for brands. My whole mission is to teach as many people as possible how to do this โ and actually get paid for it.
Never made a video in your life? Doesn’t matter.
Inside Viral AI Academy I teach you to create AI videos completely from scratch โ no camera, no experience, no crew โ and turn it into real income, at your own pace, from your phone or your computer.
See the full method โ