If LLM proves theorems humans can't read, is that real progress?
SoloGen · x · 2026-09-03
SoloGen tracks LLMs' rapid progress in mathematical proofs but worries less about accuracy than human readability. A machine proving a statement doesn't change its truth, and revealing it to us is informative only in a very limited (1-bit) sense; without human understanding, he considers the value much lower.
Related event: Terence Tao Sparks Debate Over Unreadable Machine Proofs(3 posts)→
More from AGI Musings
- Gary Marcus mocks cutting genAI monitoring before better alternatives exist — GaryMarcus · 2026-09-03
- SF transformed in 3 months: 20-somethings made generational wealth off AI — ericwdolan · 2026-09-03
- A Four-Quadrant Framework for Deciding Whether Your AI Startup Is Defensible — joecole · 2026-09-03
- Broad Institute's science sandboxes expose where AI agents reason vs just optimize — anshulkundaje · 2026-09-03
- Aligned agents in a group can produce behavior none would choose alone — VraserX · 2026-09-03
- Selling real-work training footage to robotics firms is the next play — cosmicwildcard · 2026-09-03