Researchers who use hallucinated references to face arXiv ban

2026-05-19 · Source: Machine learning : nature.com subject feeds · Field: Science & Research — Artificial Intelligence & Machine Learning, Research Methodology & Innovation, Mathematics & Computational Sciences · Depth: Fundamental Awareness, quick

Summary

The preprint server arXiv has implemented a new policy banning researchers for one year if their manuscript submissions contain references hallucinated by artificial intelligence tools or other clear indicators of unchecked generative AI usage. Following this ban, authors will only be permitted to post work already accepted by a reputable peer-reviewed venue. This measure, announced by Thomas Dietterich, chair of arXiv's computer science section, addresses the increasing volume of "AI 'slop'"—low-quality content generated by LLMs—particularly prevalent in the computer science section, which accounts for approximately half of all submissions. While scientists legitimately use LLMs for tasks like literature reviews, arXiv aims to deter authors from over-relying on and failing to verify AI outputs, a concern exacerbated by the involvement of paper mills. The policy has generated varied community feedback.

Key takeaway

For research scientists preparing manuscripts for preprint servers like arXiv, you must rigorously verify all AI-generated content, especially references. Failure to meticulously check LLM outputs for hallucinations or other clear signs of unverified AI use can result in a one-year platform ban. After this penalty, your future submissions will require prior acceptance by a reputable peer-reviewed venue, significantly impacting your ability to disseminate early research. Ensure your AI tools are used responsibly and their outputs are thoroughly validated.

Key insights

Preprint servers are implementing strict policies to combat unchecked generative AI use and maintain scientific integrity.

Principles

• Unchecked AI output erodes trust in scientific literature.
• Preprint platforms are actively combating AI-generated "slop".
• Inter-platform collaboration is crucial for content integrity.

Method

arXiv's policy imposes a one-year ban for AI-hallucinated references or unchecked LLM output, followed by a requirement for prior peer-reviewed acceptance for future submissions.

In practice

• Rigorously verify all AI-generated references and content.
• Familiarize with specific platform AI usage guidelines.
• Avoid submitting content with obvious LLM conversational artifacts.

Topics

• arXiv Policy
• Generative AI
• AI Hallucinations
• Preprint Servers
• Research Integrity
• LLM Verification

Best for: Research Scientist, AI Scientist, Tech Journalist

Related on AIssential

• arXiv implements 1-year ban for papers containing incontrovertible evidence of unchecked LLM-generated errors, such as hallucinated references or results. [N] — Authors are fully responsible for all content in their submissions, even if generated by AI.
• Hallucinated citations are polluting the scientific literature. What can be done? — AI-generated hallucinated citations are rapidly increasing, polluting academic literature with fabricated and inaccurate references.
• Every AI output is a hallucination — Large Language Models inherently produce fluent, confident outputs that are often wrong yet indistinguishable from truth, posing significant reliability issues.
• Send the arXiv AI-generated slop, get a yearlong vacation from submissions — arXiv now bans AI-generated content, imposing severe penalties on authors for non-compliance.
• Open for Contributions for "Research Paper Writing Army" – 20+ autonomous agents, 6 novelty engines, and 10 adversarial reviewers - Feel Free to contribute — Sisyphus Academica is an AI agent swarm designed to generate novel, hallucination-free research papers.
• Backlash against Arxiv's proposed 1 year ban is genuinely perplexing. [D] — ArXiv's proposed ban on AI-generated "slop" reveals deep-seated issues in academic publishing ethics and accountability.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.