SECONDLY · MARK 001

The next interface is the world itself.

AI is moving beyond chat windows and into the spaces where life happens. Secondly is building the foundation for AI that can see, hear, remember, understand context, and act across real environments.

The future of AI will not be opened in a tab. It will be present around us.

SHIFT / 002

From a tab on your screen to a presence in your space.

The reason today's AI feels limited is not the model. It's the surface. A chat window cannot see the room you're in, hear what just happened, remember last Tuesday, or know that you've walked out the door. To be useful in real life, AI has to run somewhere with senses, memory, and a body. Embodied. Ambient. Physical. On the edge.

Chat-window AI
Presence AI
Lives on a screen, in a tab.
EmbodiedLives in the room, on devices, in the environment.
Sees only what you paste in.
PhysicalSenses the environment through cameras, mics, and sensors.
Forgets between sessions.
MemoryRemembers what happened yesterday, last week, in this room.
Waits for you to start a conversation.
ProactiveNotices what changed and speaks when it matters. Stays quiet otherwise.
Streams everything to the cloud.
EdgeRuns perception locally, on the edge. Raw sensor data never leaves the device.
Acts only in the digital world: browses, fills forms, calls APIs.
AgenticActs in the digital and physical world: APIs, devices, lights, doors, calls.
STACK / 003

The presence stack.

Presence is not one feature. Embodied AI needs every layer of the stack to work together: sensing the world, understanding what matters, remembering over time, running on the edge close to the user, acting across devices and environments, and knowing when to speak, wait, or stay quiet.

CHARTER / 004

AI that is present in the physical world.