Limits and opportunities of AI reviewers: Reviewing the reviews of Nature papers

arxiv.org

1 pointmarojejian1 comment

>On a composite of all three dimensions, a reviewing agent powered by GPT-5.2 scores above each paper's top-rated human reviewer (60.0% vs. 48.2%, p = 0.009)

note this was GPT-5.2

Reply on news.ycombinator.com