Recent LLM Math Successes Suggest Formal Methods Aren't Strictly Required

littmath · x · 2026-08-02

Addressing the concern that a lack of verifiable answers limits LLMs, Ethan Mollick suggests that as models improve in formal domains, they also improve in less-verifiable ones. Developer littmath adds that recent math successes actually support this, noting that they apparently don't rely heavily on formal methods.

Related event: Wharton Prof: LLM Verifiability Concerns Overstated(2 posts)→

Original post →

More from Models

Models channel →