R0049/2026-03-31/Q001-SRC01-E01¶
Extract¶
PRISMA-trAIce (Transparent Reporting of Artificial Intelligence in Comprehensive Evidence-synthesis) is an extension to the PRISMA 2020 statement designed to ensure transparent reporting when AI is used as a methodological tool in evidence synthesis. It provides a checklist for what to report, not instructions for how to conduct the analysis. The checklist covers AI tool selection, validation steps, and transparency requirements but does not constitute a system prompt or operational methodology for an AI agent.
Relevance to Hypotheses¶
| Hypothesis | Relationship | Strength |
|---|---|---|
| H1 | Weak support — shows framework development activity but is not itself a prompt | Weak |
| H2 | Contradicts — demonstrates that structured approaches to AI in research exist | Moderate |
| H3 | Supports — exemplifies partial coverage (reporting only, not full methodology) | Strong |
Context¶
PRISMA-trAIce represents the academic community's recognition that AI use in systematic reviews requires standardized transparency frameworks. However, it operates at the reporting layer (what to disclose about AI use) rather than the operational layer (how the AI should conduct research). This is a meaningful distinction for Q001.
Notes¶
The existence of PRISMA-trAIce confirms that the academic community is actively developing frameworks for AI in research, but the gap between reporting checklists and operational system prompts remains large.