AI Capability Frontiers: Will Solved Math Conjectures Become Uninteresting?
burny_tech · x · 2026-08-02
Responding to the view that AI will quickly pick the low-hanging fruit in programmatically verifiable domains (like math and coding), burnytech disagrees. He argues that even solving top-tier conjectures—which historically required inventing whole new mathematical subfields—is still far from peak mathematics, and thus, solved conjectures will never become "uninteresting."
The original quoted tweet categorizes AI verifiability into three tiers:
- Programmatically verifiable (near-free): Games, coding, math, chip design.
- Real-world verifiable (cost/time-bounded): Sciences (biology, chemistry), physical world (robotics, aerospace), forecasting.
- Verifiable with human preference: Writing, design, comedy, persuasion.
More from AGI Musings
- First-ever paper on multi-agent swarm safety lands NeurIPS spotlight — LauraRuis · 2026-10-16
- Yoav Artzi on peer review: paper flooding still pays off as review system strains under load — yoavartzi · 2026-10-16
- Letta founder: work in fields defined by problems, not methods — lateinteraction · 2026-10-16
- Why pure mathematicians will soon flood into AI x-risk debates — Dr_Atoosa · 2026-10-16
- RSI began when model outputs augmented next-gen training data, agents just escalated it — akbirthko · 2026-10-16
- Luiza Jarovsky: with AI making careers unpredictable, 'do what you love' is the best advice left — LuizaJarovsky · 2026-10-16