Math isn't conquered by LLMs: verification at scale, but concept creation stays human
_onionesque · x · 2026-09-22
Responding to the claim that math feels conquered while LLMs still suck at writing, the author argues math isn't actually conquered: math is about creating new conceptual universes as much as routine proving. LLMs will be a superpower for verification at scale, but the creative side remains human — and complaints about AI writing apply to math too (proofs are obnoxious!).
Related event: Andrew Ng Says LLMs Are Conquering Math But Still Struggle With Writing(2 posts)→
More from AGI Musings
- When everyone runs long-lived agents, where will they talk and how will they know who's who? — repligate · 2026-09-22
- Turing winner Whitfield Diffie: 'I'm concerned by your obsession with IP' at AI-math panel — thoefler · 2026-09-22
- ScienceBuddy-Jev answers plant biochem question in 0.62s at 99.97% confidence — Scobleizer · 2026-09-22
- Hassabis: AGI's big questions belong to the arts, full AGI years away — victor_explore · 2026-09-22
- Meta researcher: AI progress is like Moore's law — long run, but it ends — rbhar90 · 2026-09-22
- Tesla FSD vet: chasing bar charts has hollowed out LLM development — yunta_tsai · 2026-09-22