An idea I buried in 2015, and finished this month

I wrote the script and drew the scenes, and I wanted to make it in clay. Then I gave up and forgot it. A month ago I remembered, and built it from scratch.

AA Abdelilah Arahal
4 min read Updated 22 September 2026

In 2015 I had an idea for a short film.

I wrote the script. I drew some of the scenes. And I wanted to make it as clay stop-motion, frame by frame, the old patient way.

Then nothing happened.

Not because I was lazy. A clay film means months of daily work, a space, lighting that never moves, and a patience I did not have then. So the file went somewhere, and was forgotten.

Ten years.

Then I remembered

A month ago I was thinking about something completely different. And the old idea jumped into my head.

So I went back to it. And built it from scratch.

And when I saw it finished, I was shocked. It was beautiful. More beautiful than I had imagined it while drawing in 2015.

What actually happened

Let me be precise, because this story usually gets told naively.

The idea did not change. It is the 2015 idea. No tool generated it or suggested it to me.

The script was already written. By hand, ten years ago.

The drawings existed. And they directed the entire visual language.

One thing changed: the cost of execution.

What needed months of manual animation became possible in a fraction of the time. That is all AI did in this story. It did not give me an idea. It removed the barrier between me and an idea I already had.

Where the time used to go

In the old method, roughly:

Where the time went on a clay film
Frame-by-frame animation70%
Building models and sets20%
Lighting and shooting7%
Writing and the idea3%
An estimate from my own experience; proportions vary by project.

Look at the last row. Three percent for the idea and the writing, the part that decides whether the film is worth making at all.

Seventy percent went to repetitive manual labour. And that is precisely the part that collapsed.

What did not change

The idea. Still the hardest thing, and still yours.

Rhythm. When to cut, how long to hold a face, when to be silent. That judgement exists in no tool.

Sound. Still needs real attention, and it is what exposes amateur work fastest.

Knowing when to stop. I can now generate fifty versions of any shot. Being able to say "that is the one" got more valuable, not less.

What this means for you

If you have a buried project: take it out and look at it today. The barrier that stopped you may no longer exist. It only occurred to me to revisit mine by accident, and I regret not doing it a year earlier.

If you study film: the part you pay years to learn has shrunk. Learn story and rhythm, because that is what remains.

If you produce: the cost of trying is near zero. Rejecting an idea because it is "too expensive to execute" has stopped being a reason.

In closing

A ten-year-old idea, finished in weeks. The idea did not change. The price of making it did.

And what I keep thinking about since that day: how many other ideas did I bury because executing them was impossible then?

Try this today: open an old folder, a file, a notebook. Find something you stopped because it was hard. And ask whether it is still hard.

Common questions

Did AI make the film?
No. The idea, the script and the drawings were mine, from 2015. What changed was only the cost of execution, the animation that consumed seventy percent of the work.
What still needs a human in a short film?
The idea, the rhythm of when to cut and how long to hold, the sound, and the ability to say "that is the one" among fifty generated versions.
Does this mean anyone can make a film now?
Anyone can make shots. A film needs a story and a rhythm, and neither got cheaper. What got cheaper is the repetitive manual part.
Where do I start if I have a postponed project?
Take it out and look at it today without nostalgia, and ask one question: the barrier that stopped me then, is it still there?
Share

No comments yet

Leave a comment

Never published. Used only if I reply to you directly.

Comments are read before they appear.

Related reading

12 min read

The best framework for building AI agents in 2026

Seventeen agent frameworks, ranked by what survives production rather than by GitHub stars. Some of the most popular names sit in the bottom tier, and one of them is a security incident waiting to happen.

Start with a short call

Fifteen minutes to understand what your team does and what you want to change. If training is not the right answer, I will say so.