CONCEPT
The Self-Reinforcing Loop
The feedback mechanism by which each user interaction strengthens the algorithm's model of the user, which produces more confirming outputs, which produce more confirming interactions — the engine of monotonic bubble contraction.
The self-reinforcing loop is the mechanism that converts personalization's initial calibration into cumulative confinement. Each interaction a user has with an algorithmic system provides a signal about her preferences. The algorithm uses the signal to refine its model. The refined model produces outputs more precisely calibrated to the user's profile. The calibrated outputs generate more engagement, which produces more signals, which refine the model further. The loop is monotonic — the bubble only contracts through ordinary use. Expansion requires a deliberate act of will, and deliberate acts of will are precisely what frictionless systems are designed to make unnecessary.
In The You On AI Field Guide
In the content filter bubble, the loop operated through clicks. Each click confirmed a predicted preference; the prediction tightened; the next offering matched more precisely; the next click confirmed again. The loop required no malicious intent and no central coordination. It emerged automatically from the optimization logic that governed every recommendation system: show users