Projected reality: augmentation made of light, rendered onto the surfaces and objects around you. PrismOS reads the room and casts the interface onto any of it, a wall, a table, a page.
No lenses, nothing to wear. PrismOS understands the space and renders light directly onto its surfaces and objects, so the augmentation is physical, shared, and part of the room itself. Everyone present sees the same thing at once, with no per-user cost.
Set a book on the desk and its details appear beside it. Walk into the kitchen and your recipe is already on the counter. Ask where your keys are and the spot lights up. PrismOS tracks every object and surface, reasons about what you're doing, and responds, wherever you move.
Only UI appears. Black stays invisible. The content feels like it's part of the environment.
The headset is a dead end. The hardware keeps getting better and people keep taking it off. The display has to move off the body and into the room: always on, shared by everyone present, nothing to charge or strap on. And the AI that makes it possible only crossed the viability line in the last twelve months: real-time scene understanding · surface mapping · on-device inference.
Two engineers from inside the two dominant approaches to AR: the headset and the glasses. We built a real-time projected reality system that tracks any object and maps the surfaces and objects around it, and it won a national gold medal, recognized by the U.S. Senate. The hard part, projecting onto arbitrary surfaces and objects with real-time tracking, already works. Now we are adding the AI scene-understanding layer that only became available this year.