当人工智能看到更热:大型语言模型气候评估中的高估偏差。

IF 3.5 2区 文学 Q1 COMMUNICATION
Tenzin Tamang, Ruilin Zheng
{"title":"当人工智能看到更热:大型语言模型气候评估中的高估偏差。","authors":"Tenzin Tamang, Ruilin Zheng","doi":"10.1177/09636625251351575","DOIUrl":null,"url":null,"abstract":"<p><p>Large language models (LLMs) have emerged as a novel form of media, capable of generating human-like text and facilitating interactive communications. However, these systems are subject to concerns regarding inherent biases, as their training on vast text corpora may encode and amplify societal biases. This study investigates overestimation bias in LLM-generated climate assessments, wherein the impacts of climate change are exaggerated relative to expert consensus. Through non-parametric statistical methods, the study compares expert ratings from the Intergovernmental Panel on Climate Change 2023 Synthesis Report with responses from GPT-family LLMs. Results indicate that LLMs systematically overestimate climate change impacts, and that this bias is more pronounced when the models are prompted in the role of a climate scientist. These findings underscore the critical need to align LLM-generated climate assessments with expert consensus to prevent misperception and foster informed public discourse.</p>","PeriodicalId":48094,"journal":{"name":"Public Understanding of Science","volume":" ","pages":"9636625251351575"},"PeriodicalIF":3.5000,"publicationDate":"2025-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"When AI sees hotter: Overestimation bias in large language model climate assessments.\",\"authors\":\"Tenzin Tamang, Ruilin Zheng\",\"doi\":\"10.1177/09636625251351575\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Large language models (LLMs) have emerged as a novel form of media, capable of generating human-like text and facilitating interactive communications. However, these systems are subject to concerns regarding inherent biases, as their training on vast text corpora may encode and amplify societal biases. This study investigates overestimation bias in LLM-generated climate assessments, wherein the impacts of climate change are exaggerated relative to expert consensus. Through non-parametric statistical methods, the study compares expert ratings from the Intergovernmental Panel on Climate Change 2023 Synthesis Report with responses from GPT-family LLMs. Results indicate that LLMs systematically overestimate climate change impacts, and that this bias is more pronounced when the models are prompted in the role of a climate scientist. These findings underscore the critical need to align LLM-generated climate assessments with expert consensus to prevent misperception and foster informed public discourse.</p>\",\"PeriodicalId\":48094,\"journal\":{\"name\":\"Public Understanding of Science\",\"volume\":\" \",\"pages\":\"9636625251351575\"},\"PeriodicalIF\":3.5000,\"publicationDate\":\"2025-07-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Public Understanding of Science\",\"FirstCategoryId\":\"98\",\"ListUrlMain\":\"https://doi.org/10.1177/09636625251351575\",\"RegionNum\":2,\"RegionCategory\":\"文学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMMUNICATION\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Public Understanding of Science","FirstCategoryId":"98","ListUrlMain":"https://doi.org/10.1177/09636625251351575","RegionNum":2,"RegionCategory":"文学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMMUNICATION","Score":null,"Total":0}
引用次数: 0

摘要

大型语言模型(llm)已经成为一种新型的媒体形式,能够生成类似人类的文本并促进互动交流。然而,这些系统受到固有偏见的影响,因为它们在大量文本语料库上的训练可能会编码和放大社会偏见。本研究调查了法学硕士产生的气候评估中的高估偏差,其中气候变化的影响相对于专家共识被夸大了。通过非参数统计方法,该研究将《政府间气候变化专门委员会2023年综合报告》中的专家评级与gpt家族法学硕士的回应进行了比较。结果表明,llm系统地高估了气候变化的影响,并且当模型以气候科学家的角色提示时,这种偏差更为明显。这些发现强调了将法学硕士产生的气候评估与专家共识相结合的迫切需要,以防止误解并促进知情的公众话语。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
When AI sees hotter: Overestimation bias in large language model climate assessments.

Large language models (LLMs) have emerged as a novel form of media, capable of generating human-like text and facilitating interactive communications. However, these systems are subject to concerns regarding inherent biases, as their training on vast text corpora may encode and amplify societal biases. This study investigates overestimation bias in LLM-generated climate assessments, wherein the impacts of climate change are exaggerated relative to expert consensus. Through non-parametric statistical methods, the study compares expert ratings from the Intergovernmental Panel on Climate Change 2023 Synthesis Report with responses from GPT-family LLMs. Results indicate that LLMs systematically overestimate climate change impacts, and that this bias is more pronounced when the models are prompted in the role of a climate scientist. These findings underscore the critical need to align LLM-generated climate assessments with expert consensus to prevent misperception and foster informed public discourse.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
7.30
自引率
9.80%
发文量
80
期刊介绍: Public Understanding of Science is a fully peer reviewed international journal covering all aspects of the inter-relationships between science (including technology and medicine) and the public. Public Understanding of Science is the only journal to cover all aspects of the inter-relationships between science (including technology and medicine) and the public. Topics Covered Include... ·surveys of public understanding and attitudes towards science and technology ·perceptions of science ·popular representations of science ·scientific and para-scientific belief systems ·science in schools
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信