AniJam / music + moving image
Let Me
Have It
An AI music video experiment.
I did not make this to demonstrate what a machine can do. I wanted to see what might happen if I stayed with an idea, worked back and forth with AI, and kept going until we reached something I would not have found alone.
01 / The experiment
Begin with a feeling. Let the form catch up.
I started with a simple question: what changes when I treat AI less like a vending machine for finished work and more like another presence in the creative room?
Let Me Have It became one answer. I let the song and the film grow together through conversation, generation, selection, rejection, and revision. Each pass gave me new material, but it also made me more certain about what belonged, what felt false, and where I wanted the idea to go next.
That is what AniJam meant for me. I brought the intent, taste, memory, direction, and final decisions. AI widened the field of possibility quickly enough to keep my creative momentum alive. The work came out of that exchange.
AI did not replace my creative process. It turned it into a conversation.
02 / The process
From pulse to picture.
I used AniJam to develop the storyboard and assemble an edit, then took the pieces somewhere else when its sequence was not the one I wanted.
Find the pulse
I started with mood, energy, and emotional direction, then built and reshaped the music until it felt like a place I wanted to enter.
Give sound a visual grammar
I translated the rhythm and tension into light, movement, framing, texture, and visual ideas I could carry through the film.
Work the storyboard
AniJam followed the storyboard when it assembled the edit. I could change the sequence manually in the timeline, so I treated every generated result as a proposal rather than a final decision.
Let another tool take a pass
I asked ChatGPT 5.6 SOL to edit all the pieces together. In my eyes, it found a better video sequence—proof that no single tool can or should be relied on to do everything.
03 / Creative collaboration
The interesting part is not frictionless creation. It is better friction.
I have learned that creative work needs resistance. An unexpected image can expose one of my weak assumptions. A strange musical turn can reveal a better direction. Sometimes a failed attempt tells me more clearly what the piece should be than an immediate success.
I find AI useful here because it can keep presenting possibilities without becoming precious about any of them. But possibility is not authorship. I still have to choose. I still have to hold the direction and say, “This is the thing—and that is not.”
For me, that is the broader promise of AI as a creative collaborator: not fewer decisions, but more interesting decisions; not one supposedly magical platform, but a studio assembled from different tools, each used for what it does best.