L-PRISMA: An Extension of PRISMA in the Era of Generative Artificial Intelligence (GenAI)
#L-PRISMA #PRISMA #generative AI #systematic review #research transparency #AI guidelines #methodology extension
📌 Key Takeaways
- L-PRISMA is an extension of the PRISMA framework designed for the generative AI era.
- It adapts existing systematic review guidelines to address challenges posed by generative AI tools.
- The extension aims to enhance transparency and reproducibility in research using AI-generated content.
- L-PRISMA provides structured reporting standards for studies involving generative AI methodologies.
📖 Full Retelling
🏷️ Themes
Research Methodology, Artificial Intelligence
Entity Intersection Graph
No entity connections available yet for this article.
Deep Analysis
Why It Matters
This development matters because it addresses a critical gap in research methodology as generative AI becomes increasingly integrated into academic and scientific work. It affects researchers, journal editors, peer reviewers, and policymakers who rely on systematic reviews and meta-analyses for evidence-based decision making. The extension ensures transparency and reproducibility in studies using AI tools, which is essential for maintaining scientific integrity. Without such guidelines, AI-generated content could undermine trust in research findings across medicine, social sciences, and other fields.
Context & Background
- PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) was first published in 2009 as an evidence-based minimum set of items for reporting systematic reviews
- The original PRISMA guidelines have been updated several times, with PRISMA 2020 being the current standard for transparent reporting
- Generative AI tools like ChatGPT, Claude, and Gemini are increasingly being used in research for literature searching, data extraction, and manuscript drafting
- There have been growing concerns about AI hallucinations, bias amplification, and lack of transparency in AI-assisted research processes
- Previous reporting guidelines like CONSORT and STROBE have also developed extensions for specific methodologies or technologies
What Happens Next
Research teams will begin adopting L-PRISMA for ongoing systematic reviews involving AI tools, with initial validation studies likely appearing within 6-12 months. Major journals will update their submission guidelines to require L-PRISMA compliance for AI-assisted reviews, potentially starting with medical and social science publications. The EQUATOR Network will likely incorporate L-PRISMA into their reporting guideline library within the next year. Training workshops and online resources for researchers will emerge to facilitate implementation.
Frequently Asked Questions
L-PRISMA addresses AI applications in literature searching, screening, data extraction, risk of bias assessment, and manuscript preparation. It provides reporting standards for how AI tools were prompted, validated, and integrated throughout the systematic review process to ensure transparency.
L-PRISMA adds specific reporting items for AI tool usage while maintaining all original PRISMA requirements. It includes new sections on AI tool selection, prompt engineering, output validation, and human-AI collaboration protocols that weren't relevant when PRISMA was originally developed.
Initially, implementation may require additional time for documentation, but the guidelines are designed to enhance efficiency through standardized reporting. Proper documentation of AI use should ultimately reduce time spent on revisions and improve review quality and reproducibility.
L-PRISMA was developed through an international Delphi consensus process involving systematic review methodologies, AI researchers, journal editors, and evidence synthesis experts. The development followed established guideline development protocols to ensure methodological rigor.
L-PRISMA is specifically for systematic reviews using generative AI tools. Traditional systematic reviews without AI involvement should continue using standard PRISMA 2020 guidelines, though journals may encourage L-PRISMA for all reviews as best practice.
The extension requires researchers to document validation procedures, prompt strategies to minimize bias, and methods for verifying AI-generated content. It mandates reporting of AI limitations and potential biases, forcing transparency about these known challenges in AI-assisted research.