GPT detectors are biased against non-native English writers

GPT detectors frequently misclassify non-native English writing as AI generated, raising concerns about fairness and robustness. Addressing the biases in these detectors is crucial to prevent the marginalization of non-native English speakers in evaluative and educational settings and to create a mo...

Descripción completa

Detalles Bibliográficos
Autores principales: Liang, Weixin, Yuksekgonul, Mert, Mao, Yining, Wu, Eric, Zou, James
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2023
Materias:
Opinion
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10382961/
https://www.ncbi.nlm.nih.gov/pubmed/37521038
http://dx.doi.org/10.1016/j.patter.2023.100779
_version_ 1785080791599939584
author Liang, Weixin
Yuksekgonul, Mert
Mao, Yining
Wu, Eric
Zou, James
author_facet Liang, Weixin
Yuksekgonul, Mert
Mao, Yining
Wu, Eric
Zou, James
author_sort Liang, Weixin
collection PubMed
description GPT detectors frequently misclassify non-native English writing as AI generated, raising concerns about fairness and robustness. Addressing the biases in these detectors is crucial to prevent the marginalization of non-native English speakers in evaluative and educational settings and to create a more equitable digital landscape.
format Online
Article
Text
id pubmed-10382961
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Elsevier
record_format MEDLINE/PubMed
spelling pubmed-103829612023-07-30 GPT detectors are biased against non-native English writers Liang, Weixin Yuksekgonul, Mert Mao, Yining Wu, Eric Zou, James Patterns (N Y) Opinion GPT detectors frequently misclassify non-native English writing as AI generated, raising concerns about fairness and robustness. Addressing the biases in these detectors is crucial to prevent the marginalization of non-native English speakers in evaluative and educational settings and to create a more equitable digital landscape. Elsevier 2023-07-10 /pmc/articles/PMC10382961/ /pubmed/37521038 http://dx.doi.org/10.1016/j.patter.2023.100779 Text en © 2023 The Author(s) https://creativecommons.org/licenses/by-nc-nd/4.0/This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
spellingShingle Opinion
Liang, Weixin
Yuksekgonul, Mert
Mao, Yining
Wu, Eric
Zou, James
GPT detectors are biased against non-native English writers
title GPT detectors are biased against non-native English writers
title_full GPT detectors are biased against non-native English writers
title_fullStr GPT detectors are biased against non-native English writers
title_full_unstemmed GPT detectors are biased against non-native English writers
title_short GPT detectors are biased against non-native English writers
title_sort gpt detectors are biased against non-native english writers
topic Opinion
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10382961/
https://www.ncbi.nlm.nih.gov/pubmed/37521038
http://dx.doi.org/10.1016/j.patter.2023.100779
work_keys_str_mv AT liangweixin gptdetectorsarebiasedagainstnonnativeenglishwriters
AT yuksekgonulmert gptdetectorsarebiasedagainstnonnativeenglishwriters
AT maoyining gptdetectorsarebiasedagainstnonnativeenglishwriters
AT wueric gptdetectorsarebiasedagainstnonnativeenglishwriters
AT zoujames gptdetectorsarebiasedagainstnonnativeenglishwriters

Ejemplares similares