Preprint server arXiv will ban submitters of AI-generated hallucinations

Ars TechnicaMay 15, 2026
arxivai-generated-contentpeer-reviewhallucinationsscientific-integrity

The preprint server arXiv is implementing strict measures against the submission of AI-generated content that includes inaccuracies and hallucinations. Submitters found to be responsible for such content will face a one-year ban and will be required to have future submissions undergo peer review. This move aims to maintain the integrity of scientific literature in light of increasing concerns over the reliability of AI-generated outputs.

Read original source
← Back to AI Policy & Regulation