<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="go-etal-2026-forewarned">
    <titleInfo>
        <title>Forewarned Is Not Forearmed: How Coordinated LLMs Shift Reader Opinions in News Comment Sections</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Wooyoung</namePart>
        <namePart type="family">Go</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yujin</namePart>
        <namePart type="family">Potter</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Hyoungshick</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Seungjin</namePart>
        <namePart type="family">Ryu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yongdae</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sunipa</namePart>
            <namePart type="family">Dev</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Jessy</namePart>
            <namePart type="family">Li</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Desmond</namePart>
            <namePart type="family">Elliott</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Large language models (LLMs) can generate human-like online comments at scale, raising concerns that coordinated AI campaigns could manipulate public opinion. While prior work has shown that single AI messages can persuade individual readers, it remains untested whether coordinated multi-agent LLM campaigns, in which multiple agents jointly plan, divide roles, and manage visibility, can shift reader opinions in news comment sections, and whether per-comment AI labels can mitigate such influence. We address these questions through a controlled experiment with 1,000 U.S. adults reading news articles on a Reddit-like platform. Each participant was randomly assigned to one of six conditions, crossing three comment sources (human, independently generated AI, or coordinated multi-agent LLM) with two label settings (per-comment “Likely AI-generated” badges from a deployed detector, shown or hidden). Coordinated LLM comments did not increase average persuasion over independently generated AI comments, but disproportionately converted readers who initially disagreed and compressed opinion diversity by 12%, manufacturing apparent consensus where genuine disagreement existed. The labels reduced the perceived believability of comments but did not reliably reduce actual opinion change. These findings reveal what we call a Knowing-Resisting dissociation between source recognition and behavioral resistance, and indicate that per-comment disclosure alone is insufficient to counter coordinated LLM opinion campaigns.</abstract>
    <identifier type="citekey">go-etal-2026-forewarned</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1862/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>42361</start>
            <end>42391</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Forewarned Is Not Forearmed: How Coordinated LLMs Shift Reader Opinions in News Comment Sections
%A Go, Wooyoung
%A Potter, Yujin
%A Kim, Hyoungshick
%A Ryu, Seungjin
%A Kim, Yongdae
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F go-etal-2026-forewarned
%X Large language models (LLMs) can generate human-like online comments at scale, raising concerns that coordinated AI campaigns could manipulate public opinion. While prior work has shown that single AI messages can persuade individual readers, it remains untested whether coordinated multi-agent LLM campaigns, in which multiple agents jointly plan, divide roles, and manage visibility, can shift reader opinions in news comment sections, and whether per-comment AI labels can mitigate such influence. We address these questions through a controlled experiment with 1,000 U.S. adults reading news articles on a Reddit-like platform. Each participant was randomly assigned to one of six conditions, crossing three comment sources (human, independently generated AI, or coordinated multi-agent LLM) with two label settings (per-comment “Likely AI-generated” badges from a deployed detector, shown or hidden). Coordinated LLM comments did not increase average persuasion over independently generated AI comments, but disproportionately converted readers who initially disagreed and compressed opinion diversity by 12%, manufacturing apparent consensus where genuine disagreement existed. The labels reduced the perceived believability of comments but did not reliably reduce actual opinion change. These findings reveal what we call a Knowing-Resisting dissociation between source recognition and behavioral resistance, and indicate that per-comment disclosure alone is insufficient to counter coordinated LLM opinion campaigns.
%U https://aclanthology.org/2026.emnlp-main.1862/
%P 42361-42391

Markdown (Informal)

[Forewarned Is Not Forearmed: How Coordinated LLMs Shift Reader Opinions in News Comment Sections](https://aclanthology.org/2026.emnlp-main.1862/) (Go et al., EMNLP 2026)

• Forewarned Is Not Forearmed: How Coordinated LLMs Shift Reader Opinions in News Comment Sections (Go et al., EMNLP 2026)

ACL

• Wooyoung Go, Yujin Potter, Hyoungshick Kim, Seungjin Ryu, and Yongdae Kim. 2026. Forewarned Is Not Forearmed: How Coordinated LLMs Shift Reader Opinions in News Comment Sections. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 42361–42391, Budapest, Hungary. Association for Computational Linguistics.