ArXiv Imposes One-Year Ban on Authors Submitting AI-Generated Papers

• May 16, 2026

ArXiv, a widely used platform for sharing scientific preprints, has announced a new policy targeting the submission of research papers generated by artificial intelligence. The service will now impose a one-year submission ban on authors found to upload low-quality content produced by neural networks.

Enforcing Quality Standards in AI-Generated Research

As AI technologies become increasingly capable of generating text, concerns over the integrity and quality of scientific publications have grown. In response, ArXiv aims to curb the dissemination of substandard research that might mislead the scientific community or clutter the repository with unreliable material.

The platform’s new rule specifically addresses the issue of AI-generated content that fails to meet expected academic standards. Authors whose submissions are identified as containing such content will be temporarily barred from submitting new work for a full year. This enforcement is intended to maintain the credibility and reliability of ArXiv as a resource for researchers worldwide.

While AI tools offer powerful assistance in drafting and editing manuscripts, the fine line between helpful automation and producing low-quality, machine-generated papers is at the core of the policy. ArXiv’s move reflects a broader debate within academic circles about appropriate use of AI technologies in research publication.

The ban emphasizes the responsibility on authors to ensure that the material they upload to the platform is original, rigorously vetted, and adheres to accepted scholarly standards. ArXiv’s decision represents a significant step toward setting boundaries for AI involvement in scientific communication and preserving the platform’s role as a trusted source for pre-publication research.

ArXiv will suspend users from submitting for a year if they upload low-quality AI-generated research papers.