Mathematician pushes back on Anthropic RH proof skepticism: using everything is a good sign

harris_edouard · x · 2026-11-22

Amid overreactions to Anthropic's Riemann hypothesis proof attempt, Jeremie Harris argues skeptics go too far, calling it a combination of ASCII symbols all present in the training set. Mathematician Alex Kontorovich counters that this actually makes him more confident the proof has a good chance: RH proofs historically require using everything, so the model combining every symbol-level trick is a promising sign, not evidence of fakery.

Original post →

More from Models

Models channel →