Related Experiment Video
Updated: Feb 17, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Optimizing the key information section of informed consent documents in biomedical research: A multi-phased study
Shahana Chumki1, Caleb Smith2, Terri Ridenour1
1Office of the Vice President for Research, University of Michigan, Ann Arbor, MI, USA.
Artificial intelligence (AI) tools can help standardize key information sections in informed consent documents (ICDs), improving consistency and efficiency. Ongoing human oversight is crucial for accuracy and regulatory compliance in research ethics.
Area of Science:
- Biomedical Ethics
- Artificial Intelligence in Healthcare
- Clinical Research Administration
Background:
- Informed consent documents (ICDs) lack standardized key information (KI) sections, complicating Institutional Review Board (IRB) reviews and participant understanding.
- This inconsistency poses challenges for ethical research conduct and regulatory compliance.
Purpose of the Study:
- To explore the utility of GPT-4o, a large language model (AI), in generating standardized KI sections for ICDs.
- To assess the accuracy, clarity, and usability of AI-generated KI content through expert evaluations.
Main Methods:
- An AI tool was developed to interpret and generate KI content from ICDs.
- Multi-phased evaluation involved IRB subject matter experts, principal investigators (PIs), and IRB reviewers assessing AI output for accuracy, differentiation, prioritization, and coherence.
- Iterative refinements were implemented based on expert feedback.
Main Results:
- AI-generated KI sections showed improved accuracy and clarity over iterative refinements.
- PIs found AI-generated content comparable to their own and expressed willingness to use the tool.
- IRB reviewers noted strengths in describing benefits and readability, but highlighted needs for improved risk descriptions.
Conclusions:
- AI tools show promise for enhancing consistency and efficiency in drafting KI sections for ICDs.
- Further refinement and human oversight are necessary for regulatory compliance and ethical standards.
- Collaboration between AI and human experts is vital for optimizing informed consent processes.
More Related Videos
Related Concept Videos
Improving Translational Accuracy
Improving Translational Accuracy
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
Clinical Trials: Overview
Bioequivalence: Overview
Biopharmaceutics and Pharmacokinetics: Overview

