Inspiration

What if the camera isn’t just observing - what if everything is watching? We wanted to push the boundaries of visual storytelling by creating a world where technology perceives, reflects and reacts. The goal: build emotion, not by filming life, but by generating it.

What it does

Everything is Watching is a full music video created entirely with text-to-video generation (using Google Veo 3). No live-action footage. The result: a surreal, immersive audio-visual experience that blends human creativity and machine imagination.

How we built it

Every shot was generated using Google Veo 3 through iterative prompting, visual direction, and detailed cinematic control. Each scene emerged from text, refined over weeks through trial, error, and precise creative direction.

Challenges we ran into

Maintaining visual consistency across scenes, achieving natural emotion in character movement, and controlling light and tone were major hurdles. It required constant fine-tuning of prompts to reach a believable, cinematic flow.

Accomplishments that we're proud of

This is one of the first completed music videos (by the time making) that was carried from concept to finish entirely within Veo 3. The final result feels emotionally resonant, with the music and visuals working together in a way that feels both human and otherworldly.

What we learned

AI filmmaking isn’t about replacing traditional methods, it’s about discovering new ones. The process taught us how to direct through language, how to sculpt emotion out of code, and how unpredictability can become part of the art.

What's next for Everything is Watching

We’re exploring a follow-up piece that expands on the same universe - pushing Veo and other tools even further into storytelling. The next step is to blend AI-driven visuals with real-world textures and soundscapes, creating an even more immersive hybrid form of cinema.

Built With

Share this project:

Updates

Submission history