If P != NP, AI-written proofs being easier to verify than to discover shouldn't surprise us
avt_im · x · 2026-09-06
A short philosophical take: if P != NP holds, we shouldn't be surprised that reverse-engineering an AI-written proof is almost infinitely easier than organically discovering a proof for a sufficiently hard result — a fitting theoretical footnote to AI theorem-proving milestones.
More from AGI Musings
- China restricts AI in classrooms: no solo AIGC use for primary schoolers, no AI answering for teachers — vista8 · 2026-09-06
- The Atlantic takes aim at 'labs': AI word policing signals loss of control, journalist argues — round · 2026-09-06
- Realization: most people can't tell — or don't care — if a video is AI — oyacaro · 2026-09-06
- McLuhan 1969 quote resurfaces: evolution shifted from biology to technology — round · 2026-09-06
- The Better AI Gets, the Harder I Feel I Must Work — jbarbier · 2026-09-06
- Robin Hanson Asks: What Share of Long-Term Altruism Should Go to AI Risk? — davidmanheim · 2026-09-06