Why Most Published Research Findings Are False

Science
https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.0020124

The landmark paper that started why there might be doubt when "discovers" are merely mirages produced by small, flexible, biased studies.

GPT-5.1 Pro estimates that 50-70% of all publications are false. Anything above 20 is seemingly alarming until you follow the incentives.

Published on November 24, 2025