Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 24, 2026

Using Learning Outcome Measures to assess Doctoral Nursing Education
10:07

Using Learning Outcome Measures to assess Doctoral Nursing Education

Published on: June 21, 2010

A Performance-Based Rubric for Generative AI use in Medical Students' Research Tasks: Development and Initial

Nino Shiukashvili1, Mariam Rochikashvili1, Vasil Kupradze1

  • 1Department of Basic Medical Sciences, Ken Walker International University, Tbilisi, Georgia.

Journal of Medical Education and Curricular Development
|May 14, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Galectin-3 and Its Relationship With Left Ventricular and Left Atrial Echocardiographic Changes in Hospitalized COVID-19 Patients.

Cureus·2026
Same author

Exploring the Impact of Professional Acting on Empathy Development in Medical Students.

MedEdPublish (2016)·2026
Same author

Pathogenesis and Pharmacotherapy of Acute Respiratory Distress Syndrome Induced by Pandemic Viral Infections: A Narrative Review.

Cureus·2026
Same author

Evaluation of the Diagnostic Performance of Interleukin-6, C-reactive Protein, and Soluble Triggering Receptor Expressed on Myeloid Cells<sup>-1</sup>, Individually and Combined, in Differentiating Candidemia from Bacteremia in Intensive Care Unit Patients.

Journal of global infectious diseases·2026
Same author

Primary healthcare providers' knowledge, attitudes, and practices regarding cancer screening recommendation and referral in Georgia, 2023.

The European journal of general practice·2025
Same author

Evaluation of GPT-4 Accuracy in the Interpretation of Medical Imaging: Potential Benefits, Limitations, and the Future.

Cureus·2025
This summary is machine-generated.

Medical students show significant gaps in recognizing AI limitations and verifying AI-generated content, crucial for patient safety. A new rubric assesses generative AI literacy, highlighting the need for improved training in AI documentation and critical evaluation.

Area of Science:

  • Medical Education
  • Artificial Intelligence in Healthcare
  • Digital Literacy

Background:

  • Generative AI is increasingly integrated into medical training.
  • Ensuring patient safety requires medical graduates to critically assess AI outputs, acknowledge AI limitations and biases, and transparently document AI use.
  • Developing methods to evaluate AI literacy in medical education is essential.

Purpose of the Study:

  • To develop and evaluate a performance-based rubric for assessing observable generative AI (Large Language Model - LLM) literacy behaviors in medical students.
  • To measure the reliability and validity of the rubric in an authentic educational setting.

Main Methods:

  • A four-domain rubric (AI Use Documentation, Prompt Generation, Verification, Integration) was developed and refined through pilot testing.
Keywords:
AI literacyArtificial intelligencemedical educationmedical studentsperformance-Based assessment

Related Experiment Videos

Last Updated: May 24, 2026

Using Learning Outcome Measures to assess Doctoral Nursing Education
10:07

Using Learning Outcome Measures to assess Doctoral Nursing Education

Published on: June 21, 2010

  • Third-year medical students' research proposals, AI chat transcripts, and AI-use disclosures were assessed using the rubric.
  • Inter-rater reliability was evaluated using the Intraclass Correlation Coefficient (ICC).
  • Psychometric properties, including floor effects, performance bands, and correlations with GPA, were examined.

    • Main Results:

    • The rubric demonstrated high inter-rater reliability across all domains.
    • Substantial floor effects were observed in AI Use Documentation and Verification domains, indicating low student performance in these areas.
    • Students scored significantly lower in Verification compared to Prompt Generation and Integration.
    • Total scores on the rubric did not significantly correlate with overall GPA.

    Conclusions:

    • The developed rubric is a reliable tool for assessing distinct LLM-use competencies in medical education.
    • Findings underscore critical gaps in students' ability to verify AI-generated information and document AI use transparently.
    • Competency guidance should emphasize AI limitation recognition and output verification to enhance patient safety.
    • Further validation and multi-site implementation of the rubric are recommended.