Models already beat human superforecasters, and that may be LLMs' biggest new power
jzlegion · x · 2026-11-05
jzlegion argues models are already beating human superforecasters at predicting the future, yet few are discussing it. He frames it historically: rulers once relied on oracles and prophets, while today we only use narrowly bounded algorithms — and with LLMs offering a real way to see into the future, forecasting may become one of the biggest net-new powers AI enables.
Related event: AI Models Now Beat Human Superforecasters at Predicting Events(2 posts)→
More from AGI Musings
- Sam Altman: the need to keep humans in control should ultimately constrain AI progress — Calvinball_24 · 2026-11-05
- Vibe coding is pricier dopamine than free short videos, argues dev — oran_ge · 2026-11-05
- Flex-RSI: a new benchmark measuring recursive self-improvement of frontier agents — _vztu · 2026-11-05
- Jensen Huang to young developers: become the world's best user of AI — rseroter · 2026-11-05
- Clement's poetic take: a simulated world only needs to behave real — c_valenzuelab · 2026-11-05
- Evolution can't copy homework: why species improve slower than iPhones — louisvarge · 2026-11-05