Debate: Using Closed AI to Peer-Review Academic Papers

danielrock · x · 2026-08-08

Academics have criticized the use of closed-source AI software for evaluating papers before publication, arguing that editors and authors cannot trust a black box and should use open-source harnesses and LLMs instead.

In response, supporters of the Refine tool argued that it generates comments rather than a black box "accept" or "reject" decision. Since humans read the comments and decide what to prioritize, no blind "trust" is required, making the open vs. closed-source debate irrelevant to this workflow.

Related event: Academics Call for Ban on Closed-Source AI in Peer Review(3 posts)→

Original post →

More from AGI Musings

AGI Musings channel →