What question did this study set out to answer?

Evaluate the semantic diversity of large language models compared to human divergent thinking capabilities.

January 23, 2026Scientific ReportsOpen Access

Divergent creativity in humans and large language models

View Full Paper

Ask AI

Bookmark

Evaluate the semantic diversity of large language models compared to human divergent thinking capabilities.
Analyzed creative outputs from state-of-the-art LLMs and a dataset of 100,000 humans.
Utilized the Divergent Association Task and various creative-writing tasks with objective scoring.
Varying linguistic strategy prompts and temperature for LLMs to assess impact on performance.
LLMs exceeded average human scores on the Divergent Association Task.
LLMs approached but did not surpass the mean creativity scores of highly creative humans.
Top performing LLMs were still outperformed by the upper half of human participants.

No takes yet. Share an insight, caveat, or question.

Evaluation reveals LLMs approach human creativity but remain inferior to top performers in divergence tasks.

Bellemare-Pepin et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark