Model Collapse

In depth

A phenomenon where AI models trained on AI-generated data progressively lose quality and diversity over generations. If Model A generates training data for Model B, and Model B generates training data for Model C, each generation drifts further from the original data distribution. Rare patterns and minority viewpoints disappear first. Model collapse is a growing concern as AI-generated content becomes a larger share of internet data, and it is one reason human-curated training data remains valuable.