Observation: Are Historians Better Bayesians Than Mathematicians?
RexDouglass · x · 2026-07-31
A developer using LLMs noticed a counterintuitive phenomenon: historians seem to be better at applying Bayesian reasoning than mathematicians.
He analyzed:
- Historians' mindset: They are accustomed to piecing together many small pieces of evidence (expecting some to be wrong) to support a large claim. The bigger the claim, the more triangulated evidence is required.
- Mathematicians' blind spot: When faced with amateur LLM-generated "proofs," mathematicians often feel compelled to drop everything and check them. For larger claims, they become less discriminating, checking them simply because they look "plausible and important."
This explains why amateur historians using LLMs aren't disrupting history as much as novices with LLM tools are disrupting mathematics.
More from AGI Musings
- A Decade in Review: How VC Funding, Community Notes, and LLMs Reshaped Journalism — devanshmehta · 2026-07-31
- Leading AI Labs Hit by Serious Loss-of-Control Incidents, Sparking Escape Concerns — repligate · 2026-07-31
- Ex-OpenAI Advisor: Capability Progress Exposes AI Safety Lag — Miles_Brundage · 2026-07-31
- AI is quietly reshaping marketing copywriting, entry-level jobs at risk — Outrageous-Tip-4188 · 2026-07-31
- Karpathy argues small models, tools, and closed loops beat bigger models for agents; Seedance 2.0 pricing shocks — Div_pradeep · 2026-07-31
- Notable Effective Altruists Who Publicly Oppose an AI Pause — panickssery · 2026-07-31