When Stereotypes GTG: The Impact of Predictive Text Suggestions on Gender Bias in Human-AI Co-Writing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Baumler, Connor, Daumé III, Hal
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917279440044032
author Baumler, Connor
Daumé III, Hal
author_facet Baumler, Connor
Daumé III, Hal
contents AI-based systems such as language models have been shown to replicate and even amplify social biases reflected in their training data. Among other questionable behaviors, this can lead to AI-generated text--and text suggestions--that contain normatively inappropriate stereotypical associations. Little is known, however, about how this behavior impacts the writing produced by people using these systems. We address this gap by measuring how much impact stereotypes or anti-stereotypes in English single-word LM predictive text suggestions have on the stories that people write using those tools in a co-writing scenario. We find that ($n=414$), LM suggestions that challenge stereotypes sometimes lead to a significantly increased rate of anti-stereotypical co-written stories. However, despite this increased rate of anti-stereotypical stories, pro-stereotypical narratives still dominated the co-written stories, demonstrating that technical debiasing is only a partially effective strategy to alleviate harms from human-AI collaboration.
format Preprint
id arxiv_https___arxiv_org_abs_2409_20390
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle When Stereotypes GTG: The Impact of Predictive Text Suggestions on Gender Bias in Human-AI Co-Writing
Baumler, Connor
Daumé III, Hal
Computation and Language
AI-based systems such as language models have been shown to replicate and even amplify social biases reflected in their training data. Among other questionable behaviors, this can lead to AI-generated text--and text suggestions--that contain normatively inappropriate stereotypical associations. Little is known, however, about how this behavior impacts the writing produced by people using these systems. We address this gap by measuring how much impact stereotypes or anti-stereotypes in English single-word LM predictive text suggestions have on the stories that people write using those tools in a co-writing scenario. We find that ($n=414$), LM suggestions that challenge stereotypes sometimes lead to a significantly increased rate of anti-stereotypical co-written stories. However, despite this increased rate of anti-stereotypical stories, pro-stereotypical narratives still dominated the co-written stories, demonstrating that technical debiasing is only a partially effective strategy to alleviate harms from human-AI collaboration.
title When Stereotypes GTG: The Impact of Predictive Text Suggestions on Gender Bias in Human-AI Co-Writing
topic Computation and Language
url https://arxiv.org/abs/2409.20390