Editorial stances on large Language models in leading nursing publications: a cross-sectional analysis

Area of Science:

• Scholarly Publishing
• Artificial Intelligence
• Nursing Research

Background:

• Scholarly publishing faces challenges integrating large language models (LLMs).
• Clear editorial standards are needed for LLM use in manuscript preparation and peer assessment.
• This study analyzes nursing publications' policies on LLMs.

Purpose of the Study:

• To analyze editorial stances of top nursing publications on LLM use.
• To identify variations in policies regarding AI in manuscript preparation and peer review.
• To assess the need for standardized LLM guidelines in nursing scholarship.

Main Methods:

• Cross-sectional analysis of the top 50 nursing journals by impact factor.
• Systematic website evaluation for LLM directives on authorship, content, images, and peer assessment.
• Extraction of journal metrics to explore correlations with policy adoption.

Main Results:

• 70% of journals have explicit LLM policies.
• 97% permit LLM content generation, but 94% prohibit LLM authorship.
• Significant divergence exists on AI-generated images (52% prohibition) and peer assessment (49% prohibition).

Conclusions:

• Nursing publications show fragmented policies on LLM integration.
• Agreement exists on AI for content generation and authorship, but not for images or peer review.
• A unified framework is necessary for ethical and responsible LLM use in nursing scholarship.