Back to AI News
arxiv

Study Finds Gender Stereotyping in LLM-Generated Children's Stories

A large-scale arXiv study examines gender stereotyping in bedtime stories generated by 16 LLMs across 35 country settings and 12 narrator roles. Using matched daughter/son prompts and a five-generation design, it analyzes 62,873 retained stories with a multi-level evaluation framework. Results show systematic narrative differences: son-directed stories emphasize higher agency and lower communion than daughter-directed stories. Hidden-state analyses also reveal decodable recipient-gender information, motivating narrative-specific bias evaluation beyond short-form benchmarks.

AI Brief

It provides a large-scale, interpretable evaluation of gender bias in an increasingly common LLM use case: generating stories for children.

arXiv:2611.24156v1 Announce Type: cross Abstract: Large language models are increasingly used to create stories for children. Since bedtime stories can shape children's perceptions of social roles, this raises concerns about whether LLMs adapt narrative portrayals according to the intended recipient's gender. We conducted a large-scale study to examine gender stereotyping in LLM-generated bedtime stories. Using matched daughter/son prompts and a five-generation design, we analyze 62,873 retained stories from 16 LLMs across 35 country settings and 12 narrator roles. We introduce a multi-level evaluation framework combining context-adjusted lexical rate contrasts, relative stereotype alignment, narrator-recipient interaction analysis, human evaluation, and representation-level interventions to examine both the consistency and interpretability of these differences. Our results reveal systematic differences in narrative framing: son-directed stories emphasize higher agency and lower communion compared to daughter-directed stories. While narrator gender expresses limited moderation on these patterns, high sociability exhibits a more pronounced interaction. Human evaluations demonstrate consistent agreement, although correspondence with lexical measures varies across narrative dimensions. Additionally, hidden-state analyses reveal decodable recipient-gender information. These findings demonstrate persistent yet context-dependent representational disparities in LLM-generated children's stories and motivate narrative-specific bias evaluation beyond short-form benchmarks.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!