aceprompt engineering skill for text-to-image AI art creation among crowdsourced participants
Source article: Prompting AI Art: An Investigation into the Creative Skill of Prompt Engineering
We are witnessing a novel era of creativity where anyone can create digital content via prompt-based learning (known as prompt engineering). This article investigates prompt engineering as a novel creative skill for creating AI art with text-to-image generation. In three consecutive studies, we explore whether crowdsourced participants can (1) discern prompt quality, (2) write prompts, and (3) refine prompts. We find that participants could evaluate prompt quality and crafted descriptive prompts, but they lacked…
Both sides are scored from claims and sources, not community votes.
G 65The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.In brief
By November 2024, researchers tested prompt engineering as a creative skill for AI art in three studies with crowdsourced participants, asking them to judge prompt quality, write prompts, and refine them.
The results matter because they show a split capability for everyday creators: basic evaluation and description is accessible, but high-quality creation is constrained by missing style-specific vocabulary, suggesting prompt engineering must be learned rather than being immediately intuitive.
Main points
- Study examined prompt engineering as a creative skill for text-to-image generation across three consecutive studies.
- Participants could evaluate prompt quality and produce descriptive prompts.
- Participants lacked style-specific vocabulary needed for effective high-quality prompting.
The gain
Crowdsourced participants were able to discern prompt quality and write descriptive prompts for text-to-image AI art generation.
The problem
Crowdsourced participants lacked the style-specific vocabulary necessary for effective prompting to achieve high-quality AI art.
The rundown
The work was structured as three consecutive studies testing whether crowdsourced participants can discern prompt quality, write prompts, and refine prompts for AI art.
Findings were interpreted as support for prompt engineering being non-intuitive, requiring practice and learning before high-quality use, with four potential futures envisioned.
Sources
- Peer-reviewedInternational Journal of Human–Computer Interaction2024-11-28
The debate