Why conspiracy theories survive: high-dimensional saddle points, per deep learning analogy

ianand · x · 2026-11-17

ianand offers an analogy: debunking conspiracy theories is hard for the same reason deep learning works. In high dimensions, almost every trap is a saddle point with one more direction that fits the data slightly better. Your relative's crazy theory survives new evidence because they just add a dimension.

Related event: Saddle Point Analogy Explains Why Conspiracy Theories Keep Adapting(2 posts)→

Original post →