The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2569
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T07:55:26.186838Z
status: published
lens: trace
sector: entertainment
headline: How prompting and editing shape psychological ownership in AI-assisted creative writing
dek: Large language models are increasingly used as writing partners, but their use complicates whether writers experience the resulting text as genuinely their own. Such psychological ownership is especially relevant in AI-assisted writing because it contributes to the immediate and extended experience of AI use, and users may sign, share, revise, or be held accountable for texts whose content was partly generated by AI. Prior work shows that greater human input can increase ownership, but it remains unclear how own…
gain_title: In a 273-person experiment, editing AI-generated stories raised writers' sense of ownership, especially after single-prompt generation.
problem_title: In a 273-person experiment, writing a story by hand gave people substantially higher ownership than any AI-assisted approach, and the gap persisted after adjusting for effort, perceived quality and motivation.
trace_subject: Psychological ownership of short stories written with AI assistance
gain_reading: In a 273-person experiment, editing AI-generated stories raised writers' sense of ownership, especially after single-prompt generation.
gain_evidence: Editing increased ownership, especially for single-prompt generation.
problem_reading: In a 273-person experiment, writing a story by hand gave people substantially higher ownership than any AI-assisted approach, and the gap persisted after adjusting for effort, perceived quality and motivation.
problem_evidence: Fully manual writing nevertheless produced substantially higher ownership than all AI-assisted conditions
quick_read: A preregistered experiment with 273 online participants asked people to write short stories either unaided or with AI help. The AI groups used one of two prompting styles, a single prompt or iterative prompting, and either edited the generated text or did not.

Fully manual writing produced substantially higher ownership than every AI-assisted condition. Editing raised ownership, most clearly for single-prompt generation, while iterative prompting helped only when editing was not available.

The gap between manual and AI-assisted writing persisted in exploratory analyses that adjusted for effort, perceived story quality and intrinsic motivation. The authors suggest that tools should avoid one-shot delegation when ownership matters and support both steering and revising.
limitation: The ownership gap analyses adjusting for effort, perceived story quality and intrinsic motivation were exploratory, and participants were online volunteers writing short stories.
tag: Dual reading
key_points: The study was a preregistered mixed factorial experiment with 273 online participants writing short stories. | Conditions were no AI, or AI with single-prompt or iterative prompting, each crossed with editing or not editing the output. | Editing increased ownership, especially for single-prompt generation. | Iterative prompting increased ownership when editing was unavailable, but not when editing was available. | The authors read this as iterative prompting and editing being partially substitutable routes to ownership. | Fully manual writing gave substantially higher ownership than all AI-assisted conditions, even after adjusting for effort, perceived quality and motivation.
rundown: Participants were online writers assigned to write short stories without AI or with AI under one of two prompting styles. Each AI style was crossed with whether participants edited the generated text, producing a mixed factorial design.

Editing raised ownership, with the largest boost after single-prompt generation. Iterative prompting raised ownership only when editing was not possible, suggesting both offer ways to exert control and invest in the text at different stages.

Despite these routes, no AI-assisted condition matched fully manual writing. Exploratory covariate analyses adjusting for effort, perceived story quality and intrinsic motivation did not close the gap. The authors conclude that writing tools should support both steering and direct revision of co-created text.
sources:
- peer_reviewed | Computers in Human Behavior: Artificial Humans | https://doi.org/10.1016/j.chbah.2026.100381 | 2026-08-01
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

abb01dc330a7fdf9dc026aebec0faebe32b81d0a783b2ed928de9d9fcdb3802c

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000