LLM-Generated Fake News Induces Truth Decay in News Ecosystem: A Case Study on Neural News Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hu, Beizhe, Sheng, Qiang, Cao, Juan, Li, Yang, Wang, Danding
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918003636961280
author Hu, Beizhe
Sheng, Qiang
Cao, Juan
Li, Yang
Wang, Danding
author_facet Hu, Beizhe
Sheng, Qiang
Cao, Juan
Li, Yang
Wang, Danding
contents Online fake news moderation now faces a new challenge brought by the malicious use of large language models (LLMs) in fake news production. Though existing works have shown LLM-generated fake news is hard to detect from an individual aspect, it remains underexplored how its large-scale release will impact the news ecosystem. In this study, we develop a simulation pipeline and a dataset with ~56k generated news of diverse types to investigate the effects of LLM-generated fake news within neural news recommendation systems. Our findings expose a truth decay phenomenon, where real news is gradually losing its advantageous position in news ranking against fake news as LLM-generated news is involved in news recommendation. We further provide an explanation about why truth decay occurs from a familiarity perspective and show the positive correlation between perplexity and news ranking. Finally, we discuss the threats of LLM-generated fake news and provide possible countermeasures. We urge stakeholders to address this emerging challenge to preserve the integrity of news ecosystems.
format Preprint
id arxiv_https___arxiv_org_abs_2504_20013
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLM-Generated Fake News Induces Truth Decay in News Ecosystem: A Case Study on Neural News Recommendation
Hu, Beizhe
Sheng, Qiang
Cao, Juan
Li, Yang
Wang, Danding
Computation and Language
Computers and Society
Information Retrieval
Online fake news moderation now faces a new challenge brought by the malicious use of large language models (LLMs) in fake news production. Though existing works have shown LLM-generated fake news is hard to detect from an individual aspect, it remains underexplored how its large-scale release will impact the news ecosystem. In this study, we develop a simulation pipeline and a dataset with ~56k generated news of diverse types to investigate the effects of LLM-generated fake news within neural news recommendation systems. Our findings expose a truth decay phenomenon, where real news is gradually losing its advantageous position in news ranking against fake news as LLM-generated news is involved in news recommendation. We further provide an explanation about why truth decay occurs from a familiarity perspective and show the positive correlation between perplexity and news ranking. Finally, we discuss the threats of LLM-generated fake news and provide possible countermeasures. We urge stakeholders to address this emerging challenge to preserve the integrity of news ecosystems.
title LLM-Generated Fake News Induces Truth Decay in News Ecosystem: A Case Study on Neural News Recommendation
topic Computation and Language
Computers and Society
Information Retrieval
url https://arxiv.org/abs/2504.20013