Nursing Students' and Faculty Experiences with Artificial Intelligence in Education: A Qualitative Study Using the Technology Acceptance Model

Synthesis: A COREQ-guided descriptive qualitative study of AI experiences in nursing education: semi-structured interviews with 28 participants (16 nursing students, 12 faculty) across two universities, analyzed deductively through the Technology Acceptance Model (TAM) with inductive extension. Four TAM-aligned categories emerged (perceived ease of use, perceived usefulness, behavioral intention, actual use). Students used AI mainly for presentations, visual content, and clinical case analysis; faculty for course materials, academic writing, and administration. Notably, AI was also described as a source of cognitive, instructional, and psychosocial support — a relational dimension TAM does not explicitly capture. Concerns included reduced critical thinking, misinformation, plagiarism, and unequal access.

Relevance to AI in Education: This study adds qualitative, discipline-specific depth to TAM-based AI adoption research in health-professions education, and surfaces an underexplored psychosocial/emotional dimension of AI use with implications for AI Literacy, student Well-Being, and equitable access.

Key Findings

• TAM holds in nursing education. Perceived ease of use, perceived usefulness, behavioral intention, and actual use shaped AI adoption, confirming TAM's continued relevance and adding task- and role-dependent nuance from qualitative data.
• Students and faculty use AI differently. Students leaned on AI for applied tasks (presentations, visual content, clinical case analysis, care planning); faculty for productivity (course materials, literature reviews, academic writing, administration). This role-based split implies tailored rather than one-size-fits-all AI education.
• AI as psychosocial support. Several participants — mainly students — described AI as providing emotional comfort during stress, a "companion" offering reassurance and a confidential space for reflection. This extends TAM's cognitive focus and, in high-stress clinical training, positions AI's usefulness partly in emotional and relational terms.
• Concerns temper positive attitudes. Misinformation (notably inaccurate pharmacological content, raising patient-safety concerns), plagiarism, diminished critical thinking, and unequal access were voiced; faculty worried over-reliance could weaken analytical reasoning and academic integrity. Participants nonetheless continued using AI — a risk-benefit tension worth further study.
• An AI-literacy and prompting gap. Few had formal AI education; participants urged training that goes beyond technical use to include critical questioning, prompt building, and verification of AI output. Faculty also emphasized institutional governance, ethical frameworks, and balancing AI with traditional teaching.
• Nursing as a leadership field. Faculty framed nurses not just as adapting to AI but as actively shaping responsible integration in healthcare's digital transformation.

What this means for practice

• Instructors. Build prompt construction and output verification directly into clinical coursework instead of teaching tool mechanics in the abstract: participants reported encountering incorrect pharmacological and clinical content, so require students to check AI-generated drug and care-plan claims against authoritative references before using them, and tie this to AI Literacy training.
• Curriculum designers. Embed AI-supported case discussion and structured, evidence-based clinical learning activities where students already use these tools — care planning, clinical case analysis, presentation development — rather than bolting on a standalone technology elective.
• Faculty developers. Train students and faculty separately: students leaned on AI for applied clinical tasks while faculty used it for course materials, literature review, and administration, so one shared training will fit neither group well.
• Administrators. Pair any AI rollout with written guidance on plagiarism, over-reliance, and unequal access — the concerns participants named — under an explicit institutional AI Governance policy rather than leaving Equity to chance.
• Instructors. Acknowledge in class that students may turn to AI for emotional reassurance during clinical stress, and pair that acknowledgment with direct referral to human support services to protect student Well-Being.

Limitations

• Both participating institutions were in the same country, which limits how broadly the findings transfer to other nursing education systems.
• The sample was small and purposively selected — 28 participants (16 nursing students, 12 faculty) recruited through maximum-variation sampling — restricting generalizability.
• Only participants who had used AI for education in the past year were included, so the study represents AI users, not non-users, and says nothing about why peers avoid the tools.
• Interviews were conducted at a single time point and rely on self-reported perceptions, leaving no view of how use changes over time and leaving room for social desirability and recall bias.

Citation

Akbaba, A., & Calik Kus, A. (2026). Nursing Students' and Faculty Experiences with Artificial Intelligence in Education: A Qualitative Study Using the Technology Acceptance Model. BMC Nursing (Article in Press).