SOMEWHERE IN TOKYO – maybe | STRATA

Inspiration

In cities where collapse is no longer shocking, how do people respond to beauty, silence, or gravity? This project began as a meditation on unnoticed disasters—a poetic study of collective numbness in an overstimulated society. Tokyo, or something like it, became the stage.

What it does

This is a silent visual poem. No dialogue, no lipsync. Just ambient sound, atmosphere, and slow gravity. It depicts a surreal urban landscape where destruction is constant, and people no longer flinch.

How we built it

The entire video was generated using Veo3. We focused on one-take sequences with subtle camera motion, avoiding cuts or artificial transitions. Every scene was composed with cinematic realism and layered urban textures, using Veo3’s prompt-to-video capabilities exclusively.

Challenges we ran into

The main challenge was generating a coherent, emotionally resonant world using only Veo3’s input prompts and motion dynamics. Without relying on dialogue, music, or cuts, we had to let timing, gravity, and silence carry the weight of the emotion.

Accomplishments that we’re proud of

We succeeded in evoking a dense atmosphere using only AI-generated video. No real cameras, no performers, no edits—just a world imagined and rendered within a single model. The result feels like a memory someone forgot to live through.

What we learned

AI-generated video requires a new kind of authorship—one based on timing, suggestion, and negative space. In Veo3, we found not just a tool, but a collaborator in emotional minimalism.

What’s next for My Awesome AI Project

Next, we’ll release the second episode of the “Somewhere in Tokyo – maybe.” series, themed around sleepwalking through cities. The focus will be on reducing the resolution of time and space, inviting viewers to experience the city as a shared hallucination. We aim to create a new kind of Tokyo—not the one we live in, but the one that quietly lives inside us.

Built With

• suno
• veo3