Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0352
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-07-20T08:57:56.995708Z
status: published
lens: g_space
sector: science
headline: Experimental narratives: A comparison of human crowdsourced storytelling and AI storytelling
dek: Abstract The paper proposes a framework that combines behavioral and computational experiments employing fictional prompts as a novel tool for investigating cultural artifacts and social biases in storytelling both by humans and generative AI. The study analyzes 250 stories authored by crowdworkers in June 2019 and 80 stories generated by GPT-3.5 and GPT-4 in March 2023 by merging methods from narratology and inferential statistics. Both crowdworkers and large language models responded to identical prompts about…
gain_title: GPT-3.5 and especially GPT-4 produced narratives that were more progressive on gender roles and sexuality than human crowdworker narratives when responding to identical prompts about creating and falling in love with an artificial human.
problem_title: (none)
trace_subject: (none)
gain_reading: GPT-3.5 and especially GPT-4 produced narratives that were more progressive on gender roles and sexuality than human crowdworker narratives when responding to identical prompts about creating and falling in love with an artificial human.
gain_evidence: more progressive in terms of gender roles and sexuality than those written by humans
problem_reading: (none)
problem_evidence: (none)
quick_read: Researchers tested a fiction-based framework that asked both humans and large language models to write stories about creating and falling in love with an artificial human. They analyzed 250 crowdworker stories from June 2019 and 80 stories from GPT-3.5 and GPT-4 from March 2023 using narratology and inferential statistics.

The comparison matters because it shows how collective cultural myths like Pygmalion persist across human and AI outputs while revealing divergent social patterns. Uncertainty remains about generalizability beyond default settings, the specific Pygmalionesque prompt, and the limited, time-separated samples.
limitation: Findings are bounded by small, time-separated samples and by use of default model settings without additional prompting.
tag: Evidence-backed gain
key_points: Study compared 250 human stories from June 2019 with 80 stories from GPT-3.5 and GPT-4 from March 2023 using identical Pygmalionesque prompts. | All solicited narratives present a scientific or technological pursuit, confirming pervasive presence of the Pygmalion myth in collective imaginary. | Framework merges methods from narratology and inferential statistics to use fictional prompts as tool for investigating cultural artifacts and social biases.
rundown: The experiment used identical prompts about creating and falling in love with an artificial human to elicit narratives from both crowdworkers and LLMs, enabling a direct controlled comparison.

Results showed a split pattern: AI stories were rated as more progressive on gender and sexuality, yet human stories retained an advantage in imaginative scenarios and rhetorical variety.
sources:
- peer_reviewed | Humanities and Social Sciences Communications | https://doi.org/10.1057/s41599-024-03868-8 | 2024-10-28
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

cd5c7d36567ab2ac459e8b3b072ea185e58e16668b8fc59998c851228c8a5b7b

previous

0000000000000000000000000000000000000000000000000000000000000000