@inproceedings{prama-islam-2025-evaluating,
title = "Evaluating Credibility and Political Bias in {LLM}s for News Outlets in {B}angladesh",
author = "Prama, Tabia Tanzin and
Islam, Md. Saiful",
editor = "Zhao, Jin and
Wang, Mingyang and
Liu, Zhu",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-srw.42/",
doi = "10.18653/v1/2025.acl-srw.42",
pages = "665--677",
ISBN = "979-8-89176-254-1",
abstract = "Large language models (LLMs) are widelyused in search engines to provide direct an-swers, while AI chatbots retrieve updated infor-mation from the web. As these systems influ-ence how billions access information, evaluat-ing the credibility of news outlets has becomecrucial. We audit nine LLMs from OpenAI,Google, and Meta to assess their ability to eval-uate the credibility and political bias of the top20 most popular news outlets in Bangladesh.While most LLMs rate the tested outlets, largermodels often refuse to rate sources due to in-sufficient information, while smaller modelsare more prone to hallucinations. We create adataset of credibility ratings and political iden-tities based on journalism experts' opinions andcompare these with LLM responses. We findstrong internal consistency in LLM credibil-ity ratings, with an average correlation coeffi-cient ({\ensuremath{\rho}}) of 0.72, but moderate alignment withexpert evaluations, with an average {\ensuremath{\rho}} of 0.45.Most LLMs (GPT-4, GPT-4o-mini, Llama 3.3,Llama-3.1-70B, Llama 3.1 8B, and Gemini 1.5Pro) in their default configurations favor theleft-leaning Bangladesh Awami League, givinghigher credibility ratings, and show misalign-ment with human experts. These findings high-light the significant role of LLMs in shapingnews and political information"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="prama-islam-2025-evaluating">
<titleInfo>
<title>Evaluating Credibility and Political Bias in LLMs for News Outlets in Bangladesh</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tabia</namePart>
<namePart type="given">Tanzin</namePart>
<namePart type="family">Prama</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Md.</namePart>
<namePart type="given">Saiful</namePart>
<namePart type="family">Islam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jin</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingyang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-254-1</identifier>
</relatedItem>
<abstract>Large language models (LLMs) are widelyused in search engines to provide direct an-swers, while AI chatbots retrieve updated infor-mation from the web. As these systems influ-ence how billions access information, evaluat-ing the credibility of news outlets has becomecrucial. We audit nine LLMs from OpenAI,Google, and Meta to assess their ability to eval-uate the credibility and political bias of the top20 most popular news outlets in Bangladesh.While most LLMs rate the tested outlets, largermodels often refuse to rate sources due to in-sufficient information, while smaller modelsare more prone to hallucinations. We create adataset of credibility ratings and political iden-tities based on journalism experts’ opinions andcompare these with LLM responses. We findstrong internal consistency in LLM credibil-ity ratings, with an average correlation coeffi-cient (\ensuremathρ) of 0.72, but moderate alignment withexpert evaluations, with an average \ensuremathρ of 0.45.Most LLMs (GPT-4, GPT-4o-mini, Llama 3.3,Llama-3.1-70B, Llama 3.1 8B, and Gemini 1.5Pro) in their default configurations favor theleft-leaning Bangladesh Awami League, givinghigher credibility ratings, and show misalign-ment with human experts. These findings high-light the significant role of LLMs in shapingnews and political information</abstract>
<identifier type="citekey">prama-islam-2025-evaluating</identifier>
<identifier type="doi">10.18653/v1/2025.acl-srw.42</identifier>
<location>
<url>https://aclanthology.org/2025.acl-srw.42/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>665</start>
<end>677</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Evaluating Credibility and Political Bias in LLMs for News Outlets in Bangladesh
%A Prama, Tabia Tanzin
%A Islam, Md. Saiful
%Y Zhao, Jin
%Y Wang, Mingyang
%Y Liu, Zhu
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-254-1
%F prama-islam-2025-evaluating
%X Large language models (LLMs) are widelyused in search engines to provide direct an-swers, while AI chatbots retrieve updated infor-mation from the web. As these systems influ-ence how billions access information, evaluat-ing the credibility of news outlets has becomecrucial. We audit nine LLMs from OpenAI,Google, and Meta to assess their ability to eval-uate the credibility and political bias of the top20 most popular news outlets in Bangladesh.While most LLMs rate the tested outlets, largermodels often refuse to rate sources due to in-sufficient information, while smaller modelsare more prone to hallucinations. We create adataset of credibility ratings and political iden-tities based on journalism experts’ opinions andcompare these with LLM responses. We findstrong internal consistency in LLM credibil-ity ratings, with an average correlation coeffi-cient (\ensuremathρ) of 0.72, but moderate alignment withexpert evaluations, with an average \ensuremathρ of 0.45.Most LLMs (GPT-4, GPT-4o-mini, Llama 3.3,Llama-3.1-70B, Llama 3.1 8B, and Gemini 1.5Pro) in their default configurations favor theleft-leaning Bangladesh Awami League, givinghigher credibility ratings, and show misalign-ment with human experts. These findings high-light the significant role of LLMs in shapingnews and political information
%R 10.18653/v1/2025.acl-srw.42
%U https://aclanthology.org/2025.acl-srw.42/
%U https://doi.org/10.18653/v1/2025.acl-srw.42
%P 665-677
Markdown (Informal)
[Evaluating Credibility and Political Bias in LLMs for News Outlets in Bangladesh](https://aclanthology.org/2025.acl-srw.42/) (Prama & Islam, ACL 2025)
ACL