Notes by ReelStack · AI-assistedUpdated 27 September 2026

How to Make AI Films Like a PRO (Advanced Course)

By prioritizing structured pre-production and split-screen reference assets over random text prompting, filmmakers can eliminate character drift and produce cohesive narrative short films.

Video thumbnail: How to Make AI Films Like a PRO (Advanced Course)
Original YouTube video

Youri van Hofwegen

Published

Watch the original video ↗

What this lesson covers

This video presents an end-to-end workflow for directing structured AI short films using pre-production planning and reference assets. Creator Youri van Hofwegen demonstrates how to lock scripts with Claude, generate consistent character and location reference sheets in OpenArt using GPT Image 2, and animate cinematic shots with Seedance 2.0.

Key takeaways from the creator

AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.

  1. 01:12 ↗

    Planning the entire story upfront in Claude ensures every shot serves a clear narrative purpose and keeps characters and locations consistent before generation.

  2. 01:45 ↗

    Designing wide shots for complex elements like ski tricks or crowds avoids AI artifacting on close-up details or complex faces.

  3. 03:22 ↗

    Generating a two-panel character sheet featuring a full-body view and facial close-up provides AI models with multiple angles to maintain subject fidelity across shots.

  4. 04:00 ↗

    Using reference images generated via high-quality image models like GPT Image 2 locks facial features and costume details prior to video generation.

  5. 05:42 ↗

    Keeping background location images empty of people prevents model hallucinations and allows crowd additions through video prompts later in the pipeline.

Workflow outlined in the video

  1. Outline narrative parameters and ask Claude to generate a structured 90-second script detailing shot sequences and twist reveals.
  2. Generate a split-panel character sheet in OpenArt using GPT Image 2 with plain backgrounds to lock costume and facial geometry.
  3. Produce clean, uninhabited environment reference images for each key location to serve as visual baselines.
  4. Review and audit character and environment asset consistency before moving into video animation.
  5. Animate environment and character references in Seedance 2.0 using shot-specific video prompts while keeping dialogue as off-screen audio.

Before you use this workflow

These notes describe the source video at its publication date. Model access, pricing, connectors and interfaces may have changed. Check the original source and the provider’s current documentation before installing an add-on, connecting an account or spending credits.

ReelStack has not independently tested this workflow. Preview one representative shot and check motion, continuity and output quality before applying it to a full production. No result, cost saving or model capability is guaranteed.

Explore the tools

How to interpret the numbers

25,720 views captured 27 September 2026. ReelStack recommendation score: 19/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.

Show the calculation
  • 45% performance: 0.41× views vs 8 other channel videos observed at a similar age. View-evidence factor 96% (views / (views + 1,000))
  • 25% freshness: 9/100 with a 14-day half-life
  • 20% momentum: 117 views/day vs 589 channel baseline (6 comparable uploads), with the same view-evidence factor
  • 10% engagement: 4/100; likes + 4× comments, smoothed with a 500-view neutral prior

Read our methodology and limitations →

Powered by ReelStack

Help keep this running

Your tip funds servers, models, and the time it takes to ship new tools faster. Set any amount below — every bit helps.