Has AI Surpassed Humans in Creative Idea Generation? A Meta-Analysis
preprint
OA: closed
CC-BY-4.0
Abstract
Has AI surpassed humans in creative idea generation? This question has gained traction as generative AI (GenAI) becomes widely used for supporting creativity. To evaluate this, we examined the first wave of experimental studies comparing human and GenAI-prompted creative idea generation, by conducting a meta-analysis of 17 studies comprising 115 effect sizes. The results showed a small but non-significant pooled effect favoring GenAI. Initial analyses suggested greater originality in GenAI ideas, but sensitivity analysis showed this was driven by a few studies with very large effects. No significant differences were found between human ideas and those generated by prompting specific GenAI models (GPT-3, GPT-3.5, GPT-4). Funnel plots and asymmetry tests indicated no evidence of publication bias, supporting the findings’ validity. This meta-analysis finds no empirical support suggesting that GenAI has surpassed humans in creative idea generation. We emphasize sociotechnical and -cultural approaches as crucial for shaping human and artificial creativity.
My notes (saved in your browser only)
Citation neighborhood (no data yet)
We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.
Source provenance
- europepmc
- last seen: 2026-05-20T01:45:00.602351+00:00
- unpaywall
- last seen: 2026-05-26T02:00:01.498150+00:00
License: CC-BY-4.0