AIEd Knowledge BaseAIEd KB
On this page

Concept

Nursing Education

created 2026-09-17 · updated 2026-10-31

Nursing Education — the preparation of nurses for licensed practice, and the subfield of health-professions education where AI is currently most studied. AI enters nursing curricula through Simulation with virtual patients and mannequins, LLM-based study and clinical-reasoning support, automated assessment, and adaptive platforms for at-risk learners. Its object is distinctive: nursing competence fuses psychomotor skill, relational practice, and professional identity formation, so a technology that raises measured performance can simultaneously erode the developmental work that produces a nurse.

Questions to Consider

  • AI-supported simulation reliably improves knowledge and Self-Efficacy but shows inconsistent, sometimes negative effects on complex psychomotor skill. Where is AI the right teacher, and where must a human body in the room remain non-negotiable?
  • Nursing students describe AI as emotional comfort during clinical stress. Is that a support to design for, or a signal that the relational apprenticeship is under-resourced?
  • AI anxiety among health-sciences students tracks job-search anxiety. Is teaching AI literacy the remedy, or does it raise the salience of a threat that institutional policy should address instead?
  • A nursing AI curriculum can be ordered by what the literature discusses or by what the clinical setting can support. When infrastructure is unreliable, should a program teach ethics and evaluation deeply and reach operational skill through low-cost simulation, rather than wait for the tools?

Introduction

Nursing education is the clinical strand of the knowledge base with the densest empirical record: system-wide reviews of AI applications, a mixed-methods synthesis of AI-powered simulation, a qualitative study of student and faculty experience, and a critical integrative review of LLM research sit alongside one another. That makes nursing a useful test case for claims the wider AI in education field makes about clinical training.

It must be kept distinct from Medical and Health Professions Education, which treats nursing as one program among medicine, pharmacy, dentistry, and allied health and organizes its account around the shared problem of scalable clinical competence. The nursing literature adds three things that page cannot carry: a competence boundary at which AI's documented benefits stop; an explicitly professional-identity framing, in which the question is who becomes a nurse rather than what a trainee can do; and a workforce layer — students' anxiety about AI displacing nursing roles, and the faculty view of nurses as shapers rather than recipients of healthcare's digital transformation.

The field also refines three adjacent concepts. From Simulation it inherits the fidelity debate but renames the deficit an "authenticity gap." From Affective Computing it takes the measurement of affect, yet its strongest claim is that affective outcomes improve while emotional depth in the interaction does not. From Self-Efficacy it takes the central mediating variable, while cautioning that confidence built in a low-risk setting may not transfer to the clinical one. Beneath all of it sits the ordinary higher education problem of workload, access, and assessment integrity, sharpened by licensure and patient consequence.

How AI appears in nursing education

  • Four application areas, recurring risks. Alrazeeni et al. (2026) reviewed 28 empirical studies (2010–April 2025) and group AI applications into personalized learning, simulation-based training, Automated Assessment, and institutional curriculum management with predictive analytics. Technological inequity, faculty preparedness gaps, and privacy and bias concerns recur. Their recommendations are concrete — embed AI simulation in emergency-care training, deploy adaptive platforms for at-risk learners, use automated tools for real-time formative Feedback, adopt diagnostic accuracy as the impact measure — within 6–12 month multi-site pilots tracking learning outcomes and trust.
  • Simulation: strong on knowledge and confidence, conditional on skill. Jiang et al. (2026)'s PRISMA-guided mixed-methods review of 19 studies (N = 1,253, mostly prelicensure students) covers generative AI/LLMs, AI-driven virtual patients and mannequins, AI-enhanced VR/mixed reality, and chatbots. The strongest designs — three RCTs plus controlled quasi-experiments — show significant gains in cognitive knowledge and affective outcomes including Self-Efficacy and communication confidence, but effects on complex psychomotor skills are inconsistent, and one RCT found AI-assisted simulation inferior to standardized-patient simulation. Qualitative meta-aggregation explains both the appeal and the limit: learners value safe, repeatable, nonjudgmental practice that bridges the theory–practice gap, but report an authenticity gap — robotic dialogue, missing nonverbal cues, no tactile examination — plus technical instability that raises extraneous cognitive load and state anxiety. The authors recommend a stepped continuum: AI for pre-learning, history-taking, and foundational reasoning; human simulators and standardized patients for complex psychomotor and emotionally loaded work; human-facilitated debriefing alongside AI feedback.
  • Acceptance is real, and relational. Akbaba and Calik Kus (2026) interviewed 28 participants (16 students, 12 faculty) and analyzed transcripts deductively through the Technology Acceptance Model. TAM held: ease of use, usefulness, intention, and use shaped adoption. The role split is practical — students used AI for presentations, visual content, clinical case analysis, and care planning; faculty for course materials, literature review, academic writing, and administration — implying tailored rather than uniform training. Two findings extend TAM's cognitive frame. Participants described AI as psychosocial support, a "companion" offering reassurance and a confidential space to reflect during clinical stress. And where AI scored higher than nurses on empathy in a high-volume context, Sun et al. (2026) read it as Structural Empathy Suppression: overwork makes authentic empathic expression unsustainable, algorithmic consistency fills the gap, and identity-constituting significance transfers to the algorithm — reframing the policy question from "how effective is this tool?" to "what conditions made it appear necessary?"
  • The open secret: practice outruns policy. Ghimire and Qiu (2025) interviewed 17 final-year students at one Australian school, 10 domestic and 7 international, and found use normalized while institutions discouraged it: "The biggest irony is that everyone knows we're using AI" (SN9). Students used it for personalized learning, to bridge theory and practice and to manage time, and international students used it for language and cultural adaptation as well. Their themes name the digital divide and student-driven demand for curriculum change, which recasts adoption as a gap between what students already do and what the program admits to.
  • Faculty read the risk as professional, not procedural. Dabkowski et al. (2026) interviewed 22 nursing academics across Australia and New Zealand and found them navigating policy that was absent, late, or written without them, sharply divergent collegial attitudes, and a line drawn at replacement rather than use. The objection was developmental before it was procedural: generating answers for a deteriorating-patient scenario means the reasoning is never practiced, and participants warned of a future cohort not fit for practice — "Copilot won't teach you to be a nurse." Assessment practice was already moving toward invigilated exams, oral vivas and process evidence, and misconduct was read as a rehearsal for unsafe clinical shortcuts, which reframes Academic Integrity as a question about who is safe to practice rather than a compliance process. The academics asked for nursing-specific guidance tied to the profession's ethical codes, GenAI literacy embedded across the curriculum, and staff development with unit co-design — not blanket prohibition.
  • Anxiety as a workforce signal. Dağ et al. (2026) surveyed 821 health-sciences students (nursing among them) and found a moderate positive correlation between AI anxiety and job-search anxiety (r = 0.233, p < 0.001), with AI anxiety a significant predictor after controlling for socio-demographics (β = 0.234, p < 0.001). AI anxiety, they argue, is not a technology attitude but a psychological factor shaping how students see their professional futures — pointing to AI Literacy and career counseling, not tooling, as the intervention.
  • Groupthink and interprofessional teams. Wiss et al. (2025) placed a generative AI agent (CALIE) into twelve newly formed interprofessional teams across seven health-professions programs, including nursing, during a 180-minute virtual problem-based learning session, prompting it to inject controversial viewpoints. Of 165 learners, 158 completed the survey; the agent was rated most useful as a tool (M = 3.49), then for feedback helpfulness (M = 3.21), and least as part of the team (M = 2.88), all pairwise differences significant (F(2, 156) = 26.01, p < .001). Facilitator stance mattered, and learners who rejected the agent's responses still used them as a socially permissible way to speak up — disagreement with the AI did team work.

What the curriculum should cover, and where the map is thin

  • A twelve-part competency map, built because none existed for nursing. Xiang et al. (2026) screened 1,750 records from PubMed, Embase, Web of Science and CNKI down to 23 included studies (2022 to 2026, 15 countries, 15 of them reviews). They mapped those studies onto the Knowledge, Attitude and Skill model, producing three dimensions and 12 sub-dimensions: foundational, technical, ethics and application knowledge; critical thinking, responsibility, ethical values, and openness and adaptability; and operational, application, collaborative and evaluation skills. They compared the existing nursing frameworks and found them untranslatable into curricula: N.U.R.S.E.S. covers roles, AI-ABCs covers objectives, CARE covers the prompt structure, and none supplies a pedagogical pathway or assessment methods.
  • Attention clusters where the writing is easy. Ethics knowledge (14 studies), responsibility (14), application skills (16) and evaluation skills (18) carry consensus; technical knowledge (9), application knowledge (9), operational skills (9) and collaboration (7) do not. That is the same boundary this page documents from outcome evidence, reached from the competency side: the field knows most about what to say about AI's ethics and least about the operational layer clinical safety depends on.
  • The remedy on offer is a sequencing rule for under-resourced programs. Teach foundational and ethics knowledge before technical knowledge; critical thinking, responsibility and ethical values before openness and adaptability; evaluation and collaboration before complex operational skill. Reach the technical layer through Simulation, and where high-fidelity or virtual reality simulation is unaffordable, through screen-based cases with pre-recorded material, open-access model demonstration and guided work with public datasets. Offline substitutes are named explicitly, including case-based discussion and conceptual demonstration of AI workflows. The stated motive is equity: if curriculum order follows the distribution of infrastructure, the concentration of the research in the United States (5 studies) and China (3) turns from a bibliometric fact into a curriculum problem.
  • The quantitative anchors are imported evidence, not results of this review. Ethics knowledge has been shown to predict AI integration (standardized coefficient 0.636, p below 0.001), and a positive perceptual future orientation to raise students' ethical values (0.308, p below 0.05). Evaluation carries the stronger clinical anchor: an AI system scored therapeutic communication with an intraclass correlation above 0.97, and a retrieval-augmented implementation reserved the final decision for the teacher's professional judgment.

Competence, identity, and the evidence gap

  • Displacement is the design variable. Sun et al. (2026)'s critical integrative review of 489 studies across 47 countries reframes LLM research in nursing around one criterion: whether the cognitive and participatory work an LLM displaces is extraneous to or constitutive of the competence being developed. Displacing documentation and routine retrieval frees working memory; displacing a full reasoning chain, an ethical justification, or an individualized care plan removes the object of learning. The evidence is not hypothetical — students using ChatGPT as a sole resource scored significantly below textbook controls on ethical standards and clinical reasoning, and an AI-integrated curriculum produced higher scores alongside less individualized, weaker-logic care plans, a performance–learning dissociation. Their Professional Identity Tension Model formalizes this across task, competency, and identity layers: outcomes measured with a tool available may index fluent performance rather than durable capability, so programs should use delayed, no-tool post-tests and transfer tasks.
  • The evidence is inverted. The same review's gap map finds no randomized or quasi-experimental study of professional identity or career development, only three of relational and ethical competency, and no follow-up beyond 12 months — while controlled evidence clusters in cognitive and technical outcomes.
  • Surveillance is documented as perception, not as improved conduct. Harerimana et al. (2026) mapped Remote Proctoring in nursing assessment across six studies (1,567 students) and found deterrence reported almost universally by proctored students (98–100% agreement in one graduate program), while the single comparative performance study found an in-person proctored cohort scoring significantly higher on a readiness exam than a remotely proctored one. Anxiety ran both ways — some students calmer at home and relieved of travel, others unable to concentrate while watched and fearful of wrongful accusation — and four of the six studies reported connectivity failure, one load shedding, making equitable access the operative constraint rather than exam security. The authors' position is that remote proctoring should be one instrument among several, governed by data-protection standards and paired with integrity-by-design and authentic assessment rather than treated as the default safeguard.
  • Readiness splits inside a single faculty. Rony et al. (2025) gathered 14 semistructured interviews and three focus groups with 16 participants, 30 nurse educators in total, at three nursing colleges in Bangladesh. Read through Technological Pedagogical Content Knowledge (TPACK), the transcripts put the readiness gap inside a single staff rather than between institutions: "Some of my colleagues are naturals with technology, but I feel like I need extra help to keep up" (P3), against unreliable connectivity and training nobody had provided. Ethics surfaced as a governance worry more than a curriculum topic, with data custody the recurring question ("Who has access to it, and how is it being used?"), and algorithmic bias the second. A mandate to adopt AI without training and infrastructure reads, in this account, as an unfunded expectation rather than a readiness failure. A published letter replying to the study makes that case at the institutional level rather than the individual one (Bridging implementation gaps: the critical role of institutional ecosystems in AI integration for nursing education).
  • A wide review base is not a deep one. El Arab et al. (2025) read 18 prior reviews and found them converging on the same three priorities: ethical and social implications, transformation of nursing education, and strategies for integration. The reviews underneath stay largely descriptive, and von Gerich et al. reported that 34 of 93 studies failed to address ethical considerations at all, so agreement at the top of the pyramid is agreement among summaries.
  • Initiative is a design variable, not a learner trait. Jin et al. (2026) randomized 79 medical and nursing students and graduates to a reactive agent that waits for prompts or a proactive one that drives sequenced questions, working over visuals generated from a four-person ward simulation. Epistemic network analysis separated the conditions with large effects on how ideas connected (Cliff's delta of -0.56 and -0.78), yet final writing scores did not differ. Their agency gap names both failure modes at once: an agent that demands more initiative than a learner has, and one that over-directs a learner who no longer needs it.
  • Risk models calibrated on another discipline flag this one almost everywhere. Kovor et al. (2026) examined privacy-preserving at-risk prediction across program-discipline partitions, and the authors' own illustration is that a nursing program might run 30-35% at-risk while a concurrent medicine program runs 15-18%. Federated logistic regression kept 99.3-99.9% of centralized performance, so the barrier to cross-program modeling is organizational rather than statistical, but the same spread means an imported threshold misfires. Their finding that forum activity ranked last in every discipline is attributed to clinical placements, where peer learning happens in person.
  • Student AI-use surveys report a correlation and call it a mechanism. Li et al. (2026) surveyed 1,020 nursing students at one institution, 98.8% of whom already used generative AI for coursework, and found AI interaction associated with educational interaction (r = 0.378) with self-regulated learning the stronger predictor (beta = 0.520). The structural model is just-identified, so it reproduces the data by construction and cannot order the three variables, and the interaction scale counts frequency and a peer norm rather than what a student does with the model.
  • Boundary conditions. The strands converge on the same restraint. AI is best evidenced for structured, repeatable objectives — foundational communication, history-taking, health education, knowledge acquisition — and least evidenced for complex psychomotor skill, emotionally loaded interaction, and long-term professional formation. Acceptance is consistently moderate to high, but acceptance of AI feedback is conditioned by trust: learners prefer human sources perceived as benevolent over AI perceived as merely competent. The reviews are also geographically narrow, dominated by uncontrolled designs and self-report, and — in the simulation review — show a near-uniform positive pattern that raises publication-bias concerns.

Common Misconceptions

"AI-supported simulation improves nursing competence across the board." The benefit stops at a documented boundary. AI-supported simulation reliably improves knowledge and Self-Efficacy but shows inconsistent, sometimes negative effects on complex psychomotor skill — and psychomotor competence is not a peripheral outcome in a profession defined by it. A program adopting simulation on the strength of knowledge gains has evidence for a different endpoint than the one licensure tests.

"The workforce anxiety around AI in nursing is really a skills gap, so AI literacy training is the remedy." The anxiety is about employment, not about competence. A survey of 821 health-sciences students found a moderate positive correlation between AI anxiety and job-search anxiety (r = 0.233, p < 0.001), with AI anxiety a significant predictor after controls. Teaching literacy addresses the fear of not being able to use a tool; it does not address the fear of not having a job — and the literature's professional-identity framing, who becomes a nurse rather than what a trainee can do, is where that question belongs.

"Learners want AI agents as collaborators." They want them as tools. Of 165 learners (158 completing), the agent was rated most useful as a tool (M = 3.49), then for feedback helpfulness (M = 3.21), and least as part of the team (M = 2.88), with all pairwise differences significant (F(2, 156) = 26.01, p < .001). Design that positions an agent as a team member is running against learner preference, not ahead of it.

"Any simulation study generalises to clinical practice." The evidence base reviewed here — 19 studies, N = 1,253, mostly prelicensure students — is largely classroom and laboratory based, and the translation from simulated competence to supervised practice is exactly the step an AI-assisted simulation does not by itself rehearse.

"AI competency for nursing means handling the tools, so training should start with them." The framework literature runs the other way. The dimensions the reviewers found consistently addressed are ethics knowledge, responsibility, application and evaluation skills; technical and operational use is where the field has least to say. A program that buys tool licenses first is building the thinnest-evidenced part of the map, and the named frameworks it might adopt instead of a curriculum, N.U.R.S.E.S., AI-ABCs and CARE, are precisely the ones without assessment methods.

Implications for Practice

For instructors

  • Do not adopt AI simulation to develop complex psychomotor skill on the strength of knowledge and self-efficacy gains. Keep deliberate supervised practice for the psychomotor endpoints and use AI where its evidence is strongest.
  • Address employment anxiety directly rather than routing it through AI literacy modules; the correlation with job-search anxiety says the concern is about the profession's future, not the learner's tool fluency.
  • Position agents as tools and feedback sources, which matches both learner preference and the design that produced the stronger ratings.

For faculty developers

  • Treat readiness as a distribution and target the lower end of it, since a mandate without training and connectivity support reproduces the gap it is meant to close.
  • Co-design the analytics with the people they describe: in one published case, nursing faculty and students built an AI-powered learning analytics dashboard together while educators kept responsibility for the judgments it informed (AI as Co-Designer and Feedback Partner: Enhancing Human Judgment).

For developers

  • Report psychomotor and knowledge outcomes separately, since they diverge in this literature.
  • Design for the tool and feedback roles first; agent-as-team-member framing is not what learners asked for.

For researchers

  • Report prelicensure and practicing-nurse samples separately, and study the transition to supervised practice if the competence claim is to mean anything clinically.

Connected Concepts

Connected Articles

Metadata

DisciplineNursing Education, Medical Education
LevelHigher Ed
Intended audienceMedical Educators, Curriculum Designers, Instructors, Researchers
Learning and instructionWorkplace Learning, Self-Efficacy
Technologies and techniquesGenerative AI, Simulation
Equity, ethics, and responsible useEquity
Foundations of AI in educationLearner Identity
Page kindSynthesis
SourceHistoryEditChat

Embed this page

Copy the code below to embed this page in a learning management system or other website. The embedded version shows only the page content — it hides the site header, navigation, and footer.