Episode Details

Even with infinite samples and a flawless learning procedure, an AI model trained recursively on its own outputs still degrades. Model collapse isn't just stasis—each generation resamples and amplifies its predecessor's errors, starving the rare tails of a distribution while the average gets richer. From Shumailov's three compounding error sources to Briesch, Sobania, and Rothlauf's logic-expression tests and a nine-generation model that turned church towers into jackrabbits, here is how iterated resampling breaks Gaussian mixtures, diffusion models, and LLMs. Key sources: • Shumailov, Shumailov et al., AI models collapse when trained on recursively generated data, *Nature* (2024), DOI 10.1038/s41586-024-07566-y — https://www.nature.com/articles/s41586-024-07566-y • Gerstgrasser, Schaeffer, Dey et al., Is Model Collapse Inevitable? Breaking the Curse of Recursion by Accumulating Real and Synthetic Data, arXiv:2404.01413v2, revised 2024-04-29 — https://arxiv.org/abs/2404.01413v2 • Wikipedia, Model collapse, last edited 2026-11-05 — https://en.wikipedia.org/wiki/Model_collapse Episode 6558 — myweirdprompts.com/n/6558 (permanent short link: myweirdprompts.com/827727) Listen on Android: MWP FM Player — myweirdprompts.com/apk