factual

AI math solutions show selection bias

The reported AI math successes suffer severe selection bias: while broadcast successes (like ~50 Erdős problems solved) look amazing, systematic studies show AI tools have only a 1-2% success rate on any given problem—they win by scale and then only the winners are publicized, so standardized challenge datasets including negative results are needed for an honest picture.

factualpending

Speaker

Terence Tao

Evidence Quote

“on any given problem an AI tool has a success rate of maybe 1% or 2%. It's just that they can buy scale, and you just pick the winners.”

Source

Dwarkesh Patel Podcast - Terence Tao – Kepler, Newton, and the true nature of mathematical discovery— Dwarkesh Patel Podcast
Created: 6/13/2026, 3:32:24 AM

Your notes

Loading notes...