We’re Different, We’re the Same: Creative Homogeneity Across LLMs

Emily Wenger, Yoed Kenett · 2025

Read We’re Different, We’re the Same: Creative Homogeneity Across LLMs by Emily Wenger, Yoed Kenett (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-19361.txt.

Description

Ask several chatbots for a story and the stories start to rhyme with each other. Emily Wenger and Yoed Kenett measure that sameness in a 2025 paper on creative homogeneity across large language models. Using standard creativity tests, they gathered responses from people and from a broad set of models, then compared how varied each group's answers were.

Model answers resembled other models' answers far more than human answers resembled each other, even after accounting for structure and length. Earlier work showed one model narrows its user's ideas; this study says the narrowing is not one brand's quirk. The models themselves converge. The authors warn that if everyone writes with similar assistants, the shared pool of ideas shrinks toward the same limited set, whichever model a person picks.

Catalog record

CREATOREmily Wenger, Yoed Kenett
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE55 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-19361

Browse the full library · About the Tower of Babel