Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Using Generative AI Personas Increases Collective Diversity in Human Ideation

Yun Wan, Yoram M Kalman

TL;DR

The paper challenges the notion that GenAI assistance invariably reduces the diversity of creative outputs by showing that introducing input diversity through ten distinct GenAI personas can preserve collective diversity in human storytelling. It replicates prior work with a richer, persona-based prompt set to generate 300 plot ideas, then evaluates diversity via semantic embeddings and NLP-based linguistic analysis. Findings indicate that while plots from the same persona are similar, across-persona plots are diverse, and human writers using these ideas maintain baseline diversity while producing richer descriptive and emotional language. The work highlights the importance of input-level diversity in human–GenAI ideation and suggests design principles for future collaborative creativity systems.

Abstract

This study challenges the widely-reported tradeoff between generative AI's (GenAI) contribution to creative outcomes and decreased diversity of these outcomes. We modified the design of such a study, by Doshi and Hauser (2024), in which participants wrote short stories either aided or unaided by GenAI plot ideas[1]. In the modified study, plot ideas were generated through ten unique GenAI "personas" with diverse traits (e.g. cultural backgrounds, thinking styles, genre preferences), creating a pool of 300 story plots. While plot ideas from any individual persona showed high similarity (average cosine similarity of 0.92), ideas across different personas exhibited substantial variation (average similarity of 0.20). When human participants wrote stories based on these diverse plot ideas, their collective outputs maintained the same level of diversity as stories written without GenAI assistance, effectively eliminating the diversity reduction observed in [1]. Traditional text analytics further revealed that GenAI-assisted stories featured greater diversity in descriptive and emotional language compared to purely human-generated stories without GenAI assistance. Our findings demonstrate that introducing diversity at the AI input stage through distinct personas can preserve and potentially enhance the collective diversity of human creative outputs when collaborating with GenAI.

Using Generative AI Personas Increases Collective Diversity in Human Ideation

TL;DR

The paper challenges the notion that GenAI assistance invariably reduces the diversity of creative outputs by showing that introducing input diversity through ten distinct GenAI personas can preserve collective diversity in human storytelling. It replicates prior work with a richer, persona-based prompt set to generate 300 plot ideas, then evaluates diversity via semantic embeddings and NLP-based linguistic analysis. Findings indicate that while plots from the same persona are similar, across-persona plots are diverse, and human writers using these ideas maintain baseline diversity while producing richer descriptive and emotional language. The work highlights the importance of input-level diversity in human–GenAI ideation and suggests design principles for future collaborative creativity systems.

Abstract

This study challenges the widely-reported tradeoff between generative AI's (GenAI) contribution to creative outcomes and decreased diversity of these outcomes. We modified the design of such a study, by Doshi and Hauser (2024), in which participants wrote short stories either aided or unaided by GenAI plot ideas[1]. In the modified study, plot ideas were generated through ten unique GenAI "personas" with diverse traits (e.g. cultural backgrounds, thinking styles, genre preferences), creating a pool of 300 story plots. While plot ideas from any individual persona showed high similarity (average cosine similarity of 0.92), ideas across different personas exhibited substantial variation (average similarity of 0.20). When human participants wrote stories based on these diverse plot ideas, their collective outputs maintained the same level of diversity as stories written without GenAI assistance, effectively eliminating the diversity reduction observed in [1]. Traditional text analytics further revealed that GenAI-assisted stories featured greater diversity in descriptive and emotional language compared to purely human-generated stories without GenAI assistance. Our findings demonstrate that introducing diversity at the AI input stage through distinct personas can preserve and potentially enhance the collective diversity of human creative outputs when collaborating with GenAI.

Paper Structure

This paper contains 17 sections, 8 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Results
  3. Story plots generated by distinctive GenAI personas are highly diverse
  4. Collective diversity of stories by GenAI-aided human participants increased
  5. Discussion
  6. Materials and Methods
  7. Developing the ten diverse personas
  8. Prompting the personas for plot ideas
  9. Story writing by participants
  10. Opening page
  11. Human only (0-plot) condition
  12. Verification question that followed the 0-plot condition
  13. 1-plot condition before asking for a plot idea
  14. 1-plot condition after asking for a plot idea
  15. 5-plots condition before asking for plot ideas
  16. ...and 2 more sections

Figures (8)

  • Figure :
  • Figure :
  • Figure :
  • Figure :
  • Figure :
  • ...and 3 more figures