Manifesto
Generative AI can make remarkable images, videos, and environments. The missing layer is interaction.
Interaction & memory
Worlds that hold context and keep responding.
We’re building experiences where people can talk, act, explore, and influence what happens next. Worlds that hold context and keep responding.
What we’re exploring
01
Characters with context and goals
02
Generated worlds with persistent state
03
Real-time stories shaped by the user
04
Agents that connect models, tools, and memory
The direction
Generated environments should feel like places that continue to exist. We’re researching temporal consistency, spatial memory, and action conditioning to move closer to that.
See the experiments →The people behind the studio
Building, experimenting, and bringing people together around living worlds.