AI Feedback Enhances Community-Based Content Moderation through Engagement with Counterarguments

Saeedeh Mohammadi, Taha Yasseri · 2025

Read AI Feedback Enhances Community-Based Content Moderation through Engagement with Counterarguments by Saeedeh Mohammadi, Taha Yasseri (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-08110.txt.

Description

Saeedeh Mohammadi and Taha Yasseri test AI feedback on community fact-checking notes in a study from 2025. Community Notes on X lets users flag misleading posts together. The system suffers partisan bias and slow verification. AI help could sharpen notes before they go public.

Participants received generated feedback on their notes in three flavors: supportive, neutral, or argumentative. They then revised their notes accordingly. All feedback improved note quality in the study. Argumentative feedback, the pushback kind, delivered the largest gains. Facing a counterargument forces a writer to tighten evidence and tone. Human-AI teamwork, friction included, may beat either side alone at moderation.

Catalog record

CREATORSaeedeh Mohammadi, Taha Yasseri
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-08110

Browse the full library · About the Tower of Babel