Economist proposes AI-assisted peer review: model reads appendices, expert judges critiques
paulnovosad · x · 2026-11-09
Development economist Paul Novosad outlines what he sees as the current optimal expert paper-review workflow under a fixed time budget: read only the intro, ask a strong AI model for its top 3 critiques, discuss and cross-check against the paper, then judge which critiques are substantive. Responding to "the expert should read the whole paper," he notes nobody was reading the dozens of appendices anyway — having an AI check whether the pre-analysis plan was followed is actually a gain. His core claim: give an expert a powerful new tool and the optimal process must involve tool use, not the old routine.
More from coding & agent
- Run Claude Code on Your iPhone: Rex + Tailscale Setup Guide — leslysandra · 2026-11-09
- Ex-OpenAI RLHF lead Diogo's north star: make intelligence as boring as databases — schwentker · 2026-11-09
- Why Typesafe's Jev reports its own confidence: no automation without knowing 99% vs 1% — schwentker · 2026-11-09
- Grok Ghost demo shows autonomous computer-use agent driven by screen OCR — Aiden_Tech_Ai · 2026-11-09
- wsp writes facts directly into model weights with closed-form, gradient-free solves — kastnerkyle · 2026-11-09
- Store catalog API talks a shopping agent out of a $149 sale, no human in the loop — schwentker · 2026-11-09