If you\u2019ve tried AI video, you\u2019ve probably hit the other wall: video AI prompting.<\/p>\n\n\n\n
You type something that sounds obvious, and the result comes back slightly off. The wrong thing moves. The motion is too much. Text loses clarity. The scene drifts. Then you end up tweaking words without getting closer.<\/p>\n\n\n\n
That is not you. That is just how video models work. They need you to be specific about two things: what should move<\/strong> and what must stay locked<\/strong>.<\/p>\n\n\n\n If you are animating a finished design or mockup, you already start with a big advantage. The subject<\/strong>, the scene<\/strong>, and the style<\/strong> are already there.<\/p>\n\n\n\n So your prompt can stay short:<\/p>\n\n\n\n Example:<\/strong><\/p>\n\n\n\n Camera: locked-off camera, perfectly static, no zoom<\/em><\/p>\n\n\n\n Action:<\/em><\/p>\n\n\n\n [0-2 s] \u2014 Headline letters slide into place with a minimal bounce<\/em><\/p>\n\n\n\n [2-6 s] \u2014 Hold perfectly still, no additional motion<\/em><\/p>\n\n\n\n Context: keep kerning and alignment consistent, no warping, no drifting<\/em><\/p>\n\n\n\n SFX: soft thump on settle<\/em><\/p>\n\n\n\n Audio: short percussive beat, minimal<\/em><\/p>\n\n\n\n If you want more aesthetic words to plug into your prompt, use Kittl\u2019s AI design prompt guide<\/a> as a vocabulary bank for lighting, composition, setting, and style keywords you can copy and remix.<\/p>\n\n\n\n\n