← sources

By every measure, Opus 4.8 beats human experts as peer reviewers in a study of 124 Harvard social science papers covering a range from statistical analyses to formal models to qualitative work.

source: reddit (reddit:OpenAI)published 2026-10-23 04:37ingested 2026-10-24 02:37

retained text

Link to the paper: [https://gking.harvard.edu/peerless/](https://gking.harvard.edu/peerless/)

Seems like all of the complaints about AI generated reviews being low quality don’t stand up to scrutiny

document fetched 2026-10-24 00:28 · canonical https://reddit.com/r/OpenAI/comments/1xd4h5f/by_every_measure_opus_48_beats_human_experts_as