
AI has reached a point where creating full cinematic videos is not only possible but incredibly fast. As a huge fan of Squid Game, when I saw the Season 3 trailer drop on Netflix, I was instantly inspired. I wanted to test just how far AI had come in generating realistic, suspenseful visuals.
What you saw in my demo video was the result — a short cinematic piece built entirely with AI tools. No actors, no cameras, no expensive equipment. Just ideas, prompts, and automation.
In this post, I’ll break down exactly how I created it step by step, from image generation to animation, sound design, and editing. By the end, you’ll see how you can build cinematic AI-driven videos of your own — quickly, efficiently, and at a quality level that’s shockingly close to professional.


Inspiration: Why Squid Game?
Squid Game works so well as a creative springboard because it’s highly visual, emotional, and symbolic. From the ominous doll to the masked guards and desperate contestants, every frame is designed to feel cinematic. That makes it a perfect fit for testing AI tools in storytelling.
The short film I made didn’t use a script or narration. Instead, I focused purely on visuals and sound effects to deliver impact. Think of it like a cinematic teaser — imagery carrying the whole weight of the story.

Step 1: Generating Images With ChatGPT
Everything started with still images. Surprisingly, ChatGPT itself can create detailed visual prompts that guide image generators to produce exactly what you want.
Here’s an example of a prompt I used:
“Contestant number 456 stands frozen in shock under warm spotlights in the arena. His bloodied tracksuit and red X mark reflect despair. Background contestants watch silently from darkened tiered seating, faces blurred. The lighting is theatrical, capturing a moment of intense realization.”
With this, I was able to generate cinematic-quality images instantly. ChatGPT even inferred that I wanted a character similar to the main contestant — no need to spell it out.
But here’s the real trick: instead of writing every prompt myself, I had ChatGPT generate a full storyboard.
I asked it:
“Create a storyboard with 10 detailed prompts for a Squid Game short film. Each prompt should represent a different cinematic scene.”
In seconds, I had 10 high-quality prompts covering everything from eerie invitations to climactic moments in the arena.


Step 2: Batch Prompting for Efficiency
Generating one image at a time is fine for demos. But for cinematic projects, you need a lot of visuals fast. That’s where batch prompting comes in.
I asked ChatGPT to export the storyboard as a JSON file:
- Each object included a filename, a unique ID (001-squid, 002-squid, etc.), and the full prompt.
- I then dropped this JSON into my batch generation script.
- In under 30 seconds, I had 10 cinematic-quality images created at once.
Compared to the UI (which often defaults to medium quality), batch mode allowed me to force high-resolution, realistic renders. This step alone made a huge difference.

Step 3: Animating Images With Cling AI 2.1
Still images are powerful, but cinema demands motion. That’s where Cling AI comes in.
Cling transforms static images into dynamic video clips. Even better, the 2.1 update now supports built-in sound effects, reducing the need for manual audio editing.
Here’s how I used it:
- Simple prompts: “Create cinematic scene” or “The camera rotates around the subject.”
- Advanced prompts: “A man in a green jacket slowly slides down the tiled wall, face tense under flickering light. The camera zooms in as footsteps echo.”
For some clips, I didn’t even write prompts. I just let Cling animate the images automatically — and the results were stunningly cinematic.
Of course, AI isn’t perfect. A few outputs had distortions or blurry sections. My solution: trim the best parts and discard the rest. The beauty of AI is speed — you can generate multiple takes until you get what you need.


Step 4: Why Not Text-to-Video?
You might be wondering: why not skip images entirely and go straight from text to video?
Simple answer: control.
Right now, text-to-video models struggle with coherence. If you ask for “a man walking toward a giant piggy bank,” you might get random artifacts, distorted faces, or completely irrelevant visuals. For cinematic, story-driven content, this unpredictability ruins the flow.
That’s why I prefer the image → video workflow. It lets me carefully craft each scene visually before animating it.
Step 5: Editing Everything in CapCut
With my Cling AI clips ready, I assembled everything in CapCut.
Here’s my workflow:
- Text layers: I added a small watermark (“AI with Gary”) to prevent confusion with official Squid Game trailers. For captions, I used a serif font (Noto Serif) for clarity.
- Audio layers: I combined three sources:
- Trailer-style background music (licensed where required).
- Dialogue captions synced with Korean voice clips.
- Built-in sound effects from Cling AI, boosted by +4dB for impact.
- Clip arrangement: I placed the most recognizable Squid Game visuals (the doll, masked guards, contestant #456) at dramatic intervals to mimic the pacing of a real Netflix teaser.
Editing didn’t need to be flashy. The AI visuals were already strong — my job was simply to arrange them into a logical, suspenseful sequence.

Lessons Learned (and Why This Matters)
- AI has crossed a threshold. With tools like ChatGPT + Cling AI, you can produce cinema-quality short films with almost no traditional filmmaking skills.
- Prompts are the new directing. The better your prompts, the closer your video gets to professional-level storytelling.
- Automation saves time. Batch scripting and Make.com workflows cut days of work into minutes.
- Editing still matters. Even with AI, human judgment is essential. Knowing what to trim, what to keep, and how to pace clips is what turns raw outputs into watchable films.



Final Thoughts
What excites me most isn’t just that I recreated a Squid Game-inspired short film in hours instead of months. It’s what this means for creators everywhere.
We’re entering a new era where AI is no longer just a novelty — it’s a creative partner. With the right workflow, you can produce cinematic, high-quality content automatically, and focus on the creative decisions that matter most.
For me, this project was just the beginning. Whether it’s cinematic teasers, story-driven shorts, or even full-length AI-generated films, the possibilities are endless.
If you’re curious about AI storytelling, automation, or YouTube content creation, stick around. I’ll be sharing more breakdowns, templates, and step-by-step guides so you can do it too.
And if you enjoyed this Squid Game-inspired experiment, hit subscribe — it really motivates me to keep pushing the limits of AI filmmaking.
The future of cinema isn’t just on Netflix. It’s in your hands, powered by AI.