The Curse of Recursion / Model Collapse

Editorial record

Plain-language summary

The authors trained models on data produced by earlier model generations and observed that each generation's outputs narrow toward the mean, losing rare events and low-probability tails until later models converge to a degenerate distribution. They gave a theoretical account attributing this to compounding statistical approximation, functional expressivity, and sampling errors across generations. The finding warned that as AI-generated text saturates the web, indiscriminate training on it degrades future models, making access to genuine human-produced data increasingly valuable.