Commentary proposes data-attestation protocols to verify scientific integrity, highlighting critical safeguards against generative AI fabrication in scientific publishing.
Key Points
To examine the threats generative artificial intelligence poses to the integrity of scientific research and propose standards to verify authentic data.
Conceptual analysis assessing the risks of generative artificial intelligence in producing fabricated and synthetic research records.
Development of institutional and journal-level policy recommendations centered on formal data-attestation requirements.
Identifies generative AI as an emerging vector capable of undermining peer review through realistic, falsified experimental data.
Proposes mandatory data attestation and preserved access to raw records to distinguish authentic evidence from synthetic content.