It’s 2025, and I’ve just finished a project where I appear on screen—without
ever stepping in front of a camera. My voice, my face, my story—but
generated entirely with AI tools, crafted with purpose and design.
This is not a gimmick. It’s a Digital Twin—my avatar, speaking in
my cloned voice, walking through cinematic spaces, delivering a manifesto on
evolution, fear, and creativity.
The process? Fully mine. I wrote the script, designed the prompts, supervised the
scenes. It’s AI-powered storytelling, but with a soul. And that
soul is human.
The Why
I started this project to explore the emotional tension between fear and
transformation. AI today feels unsettling to many. It threatens
control, identity, even jobs. But it also expands what’s possible—faster creation,
multilingual communication, visual worlds you can only dream of. My goal was to show
that tension on screen: the moment where uncertainty meets empowerment.
So I built a synthetic version of myself. And then I let him talk.
Building the Voice
To create a Digital Twin that felt real, I had to start with something deeply
personal—my voice.
Using a Beyerdynamic M70 PRO X microphone, I recorded my vocal
samples and trained a model with ElevenLabs, one of the most
advanced voice cloning platforms available. The clarity was
stunning. My clone voice could speak not only English, but multiple
languages with perfect nuance. It sounded like me. It was me.
The power of voice cloning with ElevenLabs is its flexibility. It
allowed me to script lines in any language and maintain tone, emotion, even rhythm.
This made the project globally scalable, and that’s a key insight
for any business thinking ahead: AI avatars can go multilingual, instantly.
Creating the Avatar with Runway Gen-4 and Higgsfield AI
I didn’t want a cartoon. I wanted cinema.
I used Runway Gen-4, a cutting-edge generative AI video
model, to bring my avatar to life. Gen-4 could maintain facial
consistency across scenes—essential for keeping my Digital Twin recognizably
me. Every shot was prompt-engineered for emotional tone
and filmic balance.
One of the key moments? A slow-motion tracking shot of my avatar walking
through a misty nighttime pine forest, captured in symmetrical
composition with soft moonlight shafts and pastel-muted colors. It looked straight
out of a dream—or a film by Denis Villeneuve. That scene came from a prompt like
this:
“Slow-motion tracking shot of a slim man with short dark hair, black zipped
jacket and black pants, walking through a misty nighttime pine forest.
Medium shot with 50mm lens. Centered framing, pastel muted colors,
symmetrical composition, soft moonlight shafts, layered fog, painterly
cinematic aesthetic with clean depth and theatrical balance.”
My prompts were not random. They were guided by a Custom ChatGPT I
personally trained in screenwriting and filmmaking, fed with
documentation on lighting, lenses, camera styles, and narrative flow.
Filmmaking Tools + Post-Production
I shot real-world textures using an iPhone 15 Pro Max, fitted with a
Moment Tele 58mm lens and the Blackmagic Camera
App for full manual cinematic control.
I edited everything in Adobe Premiere, and graded it in
DaVinci Resolve to ensure a cinematic color
profile that unified AI-generated and real footage. This hybrid
workflow created a seamless experience between the digital and the physical.
The Real Message
I know this might feel uncanny. Seeing “yourself” speak through code can feel like
staring in a mirror too long.
But here’s the point: if AI feels unsettling, it’s because we haven’t claimed
authorship yet. I did. I made the tools work for me—not the other way
around.
This isn’t just about digital creativity. It’s about survival and evolution
in business. The companies, the creators, the designers who embrace
these tools—those who design with AI instead of running from
it—they’ll lead. Because AI is not the designer. You
are.
This is my AI Manifesto. And it speaks with my face, my words, and
yes—my cloned voice.
