A health sciences librarian has adapted a five-tier framework for responsible AI use in coursework, giving educators a practical tool to move beyond binary debates about whether to allow or ban generative AI. The framework, originally developed for nursing students, now applies across health science disciplines and emphasizes transparency, verification, and critical engagement with AI outputs.
The approach arrives as generative AI tools raise pressing questions about academic integrity and skill development in higher education. Many faculty members worry that students may rely on AI to shortcut learning, while others see potential for enhancing research and writing. The adapted framework offers a middle path-one that ties AI use directly to learning objectives.
The framework builds on a model published by Leodoro J. Labrague in the January 2026 issue of Teaching and Learning in Nursing (doi.org/10.1016/j.teln.2025.09.012). Labrague's five-tier system-ranging from restricted use to collaborative co-creation-was designed to help nursing faculty match AI expectations with assignment goals. A librarian at Hofstra University expanded the language to fit public health, physician assistant studies, and mental health counseling, and integrated a clearer role for librarians in teaching verification and information literacy.
A spectrum of AI use
The adapted framework describes five levels of AI engagement, each with specific acceptable uses, student responsibilities, and transparency statements:
- Independent Learning: No AI use allowed. Students complete reflective or experiential work entirely on their own, with a transparency statement confirming no AI tools were used.
- Guided Exploration: AI can help clarify concepts or generate examples, but all information must be verified through scholarly sources. Students paraphrase and cite properly.
- Skill Support: AI assists with grammar, tone, or structure review of original student writing. Students retain ownership of all ideas and content.
- Research Partner: AI suggests topics or summarizes research, but students must locate and cite primary studies. Verification through databases like PubMed and CINAHL is required.
- Critical Co-Creation: AI is a collaborative partner for brainstorming outlines or comparisons. Students integrate AI input with verified sources and personal analysis.
Adapting the framework for health sciences
To make the framework usable across disciplines, the librarian shifted away from nursing-specific scenarios and toward practices common in all health science programs-working with scholarly sources, interpreting evidence, and applying professional judgment. The revision also places more emphasis on verifying AI outputs against reliable databases and positions librarians as partners in teaching those skills.
For educators exploring AI for Education, such frameworks provide a structured way to integrate AI literacy into coursework. Rather than treating AI as an all-or-nothing proposition, the five-tier approach lets instructors match AI use to specific learning objectives while maintaining academic rigor.
Putting the framework into practice
The librarian has begun using the framework in several ways: in conversations with faculty to set clear AI expectations, in one-credit library courses to introduce AI as an information tool, and in one-shot information literacy sessions to help students reflect on their own AI use. It also informs an AI for Healthcare libguide that reinforces the library's role in teaching thoughtful engagement with emerging tools.
In one-on-one consultations, the framework guides students through questions about whether and how to use AI for a particular assignment, what their instructor expects, and how to verify information responsibly.
Why this matters for educators
For educators in health sciences and beyond, this framework offers a concrete method for addressing AI use without resorting to outright bans or passive acceptance. It helps students build critical evaluation skills-checking AI outputs against library databases, reflecting on when AI supports or hinders learning, and maintaining transparency about their process. As AI tools become more embedded in academic work, the ability to set clear, tiered expectations will be essential for preserving academic integrity and developing professional judgment.
Your membership also unlocks:
