MarketPR
ArXiv Imposes One-Year Ban on Unchecked AI Paper Submissions ArXiv, the preprint repository used across computer science, mathematics, and physics, said Thursday it will suspend authors for 12 months if they submit papers containing unreviewed output from large language models.
The policy was outlined by Thomas Dietterich, chair of the repository's computer science section.
Per the announcement, moderators must find "incontrovertible evidence" that authors failed to verify AI-generated text before a submission triggers the penalty.
The repository said qualifying evidence includes fabricated references, nonsensical citations, and direct copy-paste artifacts from a language model. The rule does not prohibit the use of AI tools.
Keep reading