How To Create AI Videos of Yourself with AI VFX – Full Course

Dan Kieft presents a structured, three-level tutorial on creating AI visual effects using the OpenArt platform, which offers access to multiple AI video and image models under one subscription. The course runs from beginner-friendly object replacement effects up to intermediate compositing sequences requiring light video editing, with Cling 01 serving as the primary video generation model throughout and NanoBanana Pro handling reference image creation.

At the beginner level, Kieft demonstrates swapping a pen for a flaming candle and inserting animated AI characters — including Mickey Mouse and a Xenomorph — into live-action footage. He establishes a core two-step workflow: generate a reference image with NanoBanana Pro using the desired visual, then pass both the original footage and the reference into Cling 01’s video-to-video mode. He shows directly that using a reference image produces more consistent outputs than relying on text prompts alone.

Intermediate techniques include a lightsaber activation sequence that requires generating two separately prompted video clips — hilt only, then lightsaber on — and compositing them with a mask in Premiere Pro or CapCut to create a seamless activation moment. A cinematic seasonal time-lapse effect is also covered, using environment-swap prompts applied to static footage. Kieft is candid that current AI video generation still requires editing to cover glitches and stitch shots cleanly, and that the highest-quality AI video work is always hybrid. Google Vio 3.1 is mentioned as an additional model available within OpenArt for specific use cases.

📺 Source: Dan Kieft · Published January 06, 2026
🏷️ Format: Tutorial Demo