ArXiv bans authors who upload unchecked AI-generated papers
arxiv will issue one-year bans for submissions containing 'incontrovertible evidence' of unchecked ai generation, such as hallucinated references or meta-comments.
arxiv is enforcing a strict one-year ban on authors who submit papers containing 'incontrovertible evidence' that they did not verify ai-generated content, such as hallucinated references or prompts left in the text. thomas dietterich, the platform's computer science section chair, explained that banned users must have subsequent submissions accepted by a reputable peer-reviewed venue before posting to arxiv again. the policy requires moderators to document the evidence and section chairs to confirm penalties, though authors retain the right to appeal. this move follows arxiv's earlier restriction on computer science review articles to curb the influx of ai-generated noise.