STORY · FORSKNING_

Researcher measures AI-written content on arXiv and reveals the limitations of detection methods

An analysis of AI-generated text on the preprint server arXiv shows how much artificial content is being published there, but the study also uncovers serious methodological problems with reliably detecting AI-written research.

WHY IT MATTERS

Questions about AI-written research are becoming increasingly important for scientific integrity, and this study demonstrates that we still lack adequate tools to assess the scope of the problem.

SOURCES

MACHINE-GENERATED SUMMARY This summary is written by machine from the sources below. We sort and explain — but we are a way into the field, not the final word. Check the source when something matters to you.