Is artificial intelligence more creative than humans?: ChatGPT and the Divergent Association Task
Open accessThe DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Abstract
A fundamental premise of the future of work is that AI will replace people in many cognitive and physical tasks, leaving creativity as a core, human 21st century skill. However, the recent launch of generative AI (especially ChatGPT) has seen many claims that AI is creative. If true, then the foundation of future human work, and education, is under threat. To examine claims of AI creativity, this research applied a test of verbal divergent thinking – the Divergent Association Task – to two versions of ChatGPT (GPT3.5 and GPT4). The results are reported and compared to a large human baseline. While both forms of ChatGPT show a capacity for verbal divergent production that exceeds human means, a range of factors call into question the “creativity” of generative AI. LIFT Learning: Could AI replicate your creativity? Engage with Professor David Cropley as he discusses the rise of ChatGPT and its ability to “think” creatively at this article’s companion LIFT Learning site. Professor Cropley explains the process of divergent thinking as a measure of creativity and outlines why your creative role is probably still safe for now. The LIFT Learning site is available at https://lift.c3l.ai/courses/course-v1:LEARNINGLETTERS+0113+2023
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The future of employment: How susceptible are jobs to computerisation?
Frey, C. B., & Osborne, M. A. (2017) · Technological Forecasting and Social Change · doi:10.1016/j.techfore.2016.08.019 · Cited by 8,727 · cited 1×
“Even only recently (e.g., Cropley et al., 2022; Frey & Osborne, 2017) we were reassured that AI would replace humans only in jobs that were characterised by predictable, algorithmic cognitive and physical labour (e.g., bookkeeping, editing) leaving humans free to focus on unpredictable,…”
BackgroundGuilford, J. P. (1950) · American Psychologist · doi:10.1037/h0063487 · Cited by 3,588
Automating creativity assessment with SemDis: An open platform for computing semantic distance
Beaty, R. E., & Johnson, D. R. (2021) · Behavior Research Methods · doi:10.3758/s13428-020-01453-w · Cited by 341 · cited 1×
“In particular, Olson et al. (2021) have developed the Divergent Association Task (DAT), using the concept of semantic distance (see Beaty & Johnson, 2021), to create a robust, automated test of verbal divergent thinking that is ideally suited to ChatGPT.”
MethodNaming unrelated words predicts creativity
Olson, J. A., Nahas, J., Chmoulevitch, D., Cropper, S. J., & Webb, M. E. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2022340118 · Cited by 182 · cited 6×
“…words was entered into the DAT website (https://www.datcreativity.com/task) which returned two values: (a) the raw score (ranging from 0 to 200) and (b) the percentile rank of that score, comparing the raw score to the large sample of individuals who have completed the DAT (see Olson et al., 2021).”
BackgroundMethodResultInfluential
Cited by
How Does Narrow AI Impact Human Creativity?
Marrone, R., Cropley, D., & Medeiros, K. (2026) · Creativity Research Journal · doi:10.1080/10400419.2024.2378264 · Cited by 65
Self-citationTang, M., Hofreiter, S., Werner, C. H., Zielińska, A., & Karwowski, M. (2025) · The Journal of Creative Behavior · doi:10.1002/jocb.1519 · Cited by 32
Human‐AI Co‐Creativity: Does ChatGPT Make Us More Creative?
Medeiros, K., Cropley, D. H., Marrone, R. L., & Reiter‐Palmon, R. (2025) · The Journal of Creative Behavior · doi:10.1002/jocb.70022 · Cited by 30
Self-citationCyber-Creativity: A Decalogue of Research Challenges
Corazza, G. E., Agnoli, S., Jorge Artigau, A., Beghetto, R. A., Bonnardel, N., Coletto, I., Faiella, A., Gerardini, K., Gilhooly, K., Glăveanu, V. P., et al. (2025) · Journal of Intelligence · doi:10.3390/jintelligence13080103 · Cited by 14
“Several authors agree that understanding cyber-creativity is a crucial and urgent challenge (Binz et al. 2023; Cropley and Cropley 2023; Gl ă veanu and de Saint-Laurent 2023; Marrone et al. 2024; Rafner et al. 2023; Vinchon et al. 2023).”
Metadata
- Authors
- Cropley, D.
- Venue
- Learning Letters
- Publisher
- Adelaide University
- Published
- 2023-09-14
- Volume
- 2
- Pages
- 13
- ISSN
- 2981-877X (ISSN-L)
- OpenAlex
- W4386791285
- Cited by
- 66
- References
- 4
- FWCI
- 15.09
- Type
- journal article
- Language
- en
- Open access
- diamond
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Is artificial intelligence more creative than humans?: ChatGPT and the Divergent Association Task. https://app.cochat.ai/papers/10.59453/ll.v2.13
- The paper
- Cropley, D. (2023). Is artificial intelligence more creative than humans?: ChatGPT and the Divergent Association Task. Learning Letters. https://doi.org/10.59453/ll.v2.13
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.