Digital twin

A digital twin is a real-time computational model of a physical system — a machine, a building, a city, or a body — that mirrors the state, behavior, and history of its physical counterpart through streams of sensor data and simulation. Unlike a static CAD model or a database record, the digital twin evolves in parallel with the physical entity, enabling prediction, optimization, and intervention at a distance. The concept originates from aerospace manufacturing, where NASA used paired spacecraft simulations during the Apollo missions, but its modern form relies on the convergence of IoT sensor networks, cloud computing, and machine learning.

The epistemological stakes are high: a digital twin is not merely a representation but an operational extension of the physical system. When an AR overlay persists across sessions because it is anchored to a digital twin rather than to transient sensor readings, the twin becomes part of the user's perceptual infrastructure. The boundary between model and reality blurs, raising questions about whether the twin or the original possesses greater operational reality in contexts where decisions are made on the model rather than the thing itself.