INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9211
www.rsisinternational.org
Educational Strategies for Enhancing AI Literacy among Nursing
Students: A Systematic Review
Siti fatimah Md Shariff*, Joemmaicca Augustta anak Joggery, Noor Siah Abdul Aziz
Faculty of Technology and Applied Sciences, Open University Malaysia, Kelana Centre Point, Jalan SS
7/19 / Ss7,47301, Petaling Jaya, Selangor.
*Corresponding Author
DOI: https://dx.doi.org/10.47772/IJRISS.2025.903SEDU0698
Received: 15 November 2025; Accepted: 24 November 2025; Published: 28 November 2025
ABSTRACT
Background: Artificial intelligence (AI) is rapidly entering nursing education and practice, yet AI literacy among
nursing students remains uneven, with gaps in foundational knowledge, ethical reasoning, and applied clinical
judgment.
Objective: To synthesize educational strategies that enhance AI literacy among nursing students and identify
outcome patterns, barriers, and implementation enablers to inform curriculum, assessment, and policy.
Methods: A systematic literature review (2015–2025) was conducted across PubMed, Scopus, ScienceDirect,
CINAHL, and ERIC. English-language, peer-reviewed studies focusing on pre- or post-licensure nursing
students and reporting AI-related educational outcomes were included. Two reviewers independently screened
records, extracted data into a structured matrix, and appraised quality using CASP (qualitative) and JBI
(quantitative/mixed-methods) tools. Narrative and thematic synthesis was used to integrate findings.
Results: Of 364 records identified, 47 duplicates were removed; 317 titles/abstracts were screened, 63 full texts
were sought (60 assessed), and 28 studies met inclusion. Four pedagogical themes emerged: (1) simulation-
based learning using AI-enabled or GenAI-supported scenarios; (2) online/blended modules for scalable
foundational concepts; (3) problem-/case-based learning (PBL/CBL) to situate AI within clinical reasoning and
communication; and (4) cross-disciplinary/policy approaches aligning competencies, assessment, and
governance. Across diverse settings, interventions improved knowledge, confidence/readiness, and higher-order
thinking. Transfer to clinical judgment was strongest for PBL/CBL and simulation with structured debriefs.
Recurrent barriers included limited faculty readiness, student anxiety about AI’s impact on nursing identity,
infrastructural constraints (devices/connectivity), and uneven treatment of ethics, bias, and accountability.
Studies rarely measured longitudinal behavior change or patient-centered outcomes.
Conclusions: AI literacy is achievable at scale when foundations are delivered via blended learning, transfer is
secured through PBL/CBL, and safe practice is consolidated through simulation and guided debriefing—
underpinned by robust assessment, faculty development, equitable infrastructure, and clear policy for human-in-
the-loop accountability. Future research should adopt standardized measures and longitudinal designs to link
classroom gains to clinical behaviors and patient outcomes.
Keywords: nursing education; artificial intelligence; AI literacy; simulation; blended learning; problem-based
learning; curriculum; ethics; faculty development; systematic review.
INTRODUCTION
Artificial intelligence (AI) is rapidly reshaping healthcare by augmenting clinical reasoning, diagnostics,
workflow efficiency, and patient safety. As the largest segment of the global health workforce, nurses
increasingly encounter AI-enabled tools—from risk prediction models and decision-support systems to
conversational agents—within academic, clinical, and community settings. These shifts make AI literacy a core
competency for contemporary nursing practice. Yet, evidence from nursing education reveals uneven readiness:
many students report curiosity and positive attitudes toward AI, but demonstrate limited knowledge, inconsistent
skills, and uncertainty about appropriate, ethical use. At the same time, faculty cite gaps in expertise, curricular
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9212
www.rsisinternational.org
time, and institutional support, while persistent infrastructure constraints risk widening digital inequities between
well-resourced and resource-limited contexts. Collectively, these factors underscore an urgent need for coherent
educational strategies that build practical competence, critical judgment, and ethical awareness in AI use for
nursing.
Within this context, AI literacy in nursing extends beyond basic digital familiarity. It involves understanding
foundational concepts (e.g., data, models, and outputs), interpreting and communicating AI-derived insights,
appraising strengths and limitations (including bias, fairness, and transparency), and integrating AI responsibly
into care and learning environments. Preparing nursing students therefore requires intentional alignment of
pedagogy, assessment, faculty development, and policy frameworks. Early integration across the curriculum,
scaffolded practice opportunities, and explicit attention to ethical, legal, and social implications can help ensure
that AI complements rather than displaces humanistic, patient-centred care.
This systematic review synthesizes educational strategies designed to enhance AI literacy among nursing
students. Guided by PRISMA 2020 standards, it searched five databases (PubMed, Scopus, ScienceDirect,
CINAHL, ERIC) for English-language, peer-reviewed studies from 2015 to 2025 and included qualitative,
quantitative, and mixed-methods designs focused on pre- or post-licensure nursing students. Two independent
reviewers screened and extracted data, with study quality appraised using CASP (qualitative) and JBI
(quantitative/mixed-methods) tools. Narrative and thematic synthesis organized interventions into four broad
categories: (1) simulation-based learning that affords safe, experiential engagement with AI-supported clinical
scenarios; (2) online and blended learning that scales access and supports flexible, iterative practice; (3) problem-
/case-based learning (PBL/CBL) that situates AI concepts within clinical reasoning; and (4) cross-disciplinary
and policy approaches that embed competencies, standards, and governance considerations within curricula.
Across the included studies, simulation and blended formats consistently improved knowledge, confidence, and
higher-order thinking related to AI, while PBL/CBL deepened transfer to clinical decision-making. Nonetheless,
four cross-cutting barriers recurred: limited faculty readiness, student anxiety about the role of AI in nursing,
infrastructure constraints (especially in low-resource settings), and insufficient coverage of ethics, bias, and
accountability. Addressing these challenges requires multilevel action: targeted faculty development, structured
mentorship for students, institutional investment in digital infrastructure, and policy-aligned curricula that
integrate ethics throughout.
By mapping effective pedagogies, common barriers, and enabling conditions, this review offers a consolidated
evidence base to guide curriculum design, educator capacity building, and policy alignment. Ultimately,
strengthening AI literacy is essential to equip future nurses to partner with intelligent systems safely and
equitably, supporting sound clinical judgment, compassionate care, and better health outcomes.
Accordingly, this systematic review had four objectives: (1) to identify and synthesize educational strategies that
have been used to enhance AI literacy among pre- and post-licensure nursing students; (2) to examine patterns
in AI-related learning outcomes reported in these interventions, including knowledge, confidence/readiness,
higher-order thinking, and transfer to clinical judgment; (3) to identify recurrent barriers, facilitators, and
contextual factors that influence the implementation of AI literacy strategies in nursing education; and (4) to
derive implications for curriculum design, assessment, faculty development, and policy alignment to support
safe, ethical, and equitable integration of AI in nursing education.
METHODOLOGY
This systematic literature review followed PRISMA 2020 to ensure rigor, transparency, and replicability,
adopting a systematic review (SLR) approach to synthesize evidence on educational strategies that enhance AI
literacy among nursing students. Comprehensive searches were conducted in PubMed, Scopus, ScienceDirect,
CINAHL, and ERIC for English-language, peer-reviewed studies published from 2015 to 2025, using Boolean
operators and controlled vocabulary (e.g., MeSH) across AI-, education-, and competency-related terms.
Example strings combined concepts such as “artificial intelligence,” “machine learning,” “nursing education,”
“nursing students,” “literacy,” and “curriculum” to maximize coverage.
Inclusion criteria comprised empirical qualitative, quantitative, or mixed-methods studies focusing on pre- or
post-licensure nursing students and examining educational strategies for AI literacy or related competencies.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9213
www.rsisinternational.org
Exclusion criteria removed non-nursing populations, studies lacking educational outcomes, purely technical AI
papers, editorials/opinions/grey literature, and studies published before 2015.
Records were exported to Zotero, duplicates removed, and titles/abstracts independently screened by two
reviewers. Full texts were retrieved for potentially eligible studies, with disagreements resolved by consensus.
In total, 364 records were identified; after removing 47 duplicates, 317 remained for screening, from which 254
were excluded at title/abstract level. Sixty-three full texts were sought (three not retrievable), 60 were assessed,
and 32 were excluded for predefined reasons (non-nursing populations, no educational intervention, clinical AI
only, or duplicate conference versions), yielding 28 studies for synthesis.
Methodological quality was appraised using CASP for qualitative studies and JBI tools for quantitative and
mixed-methods designs; no study was excluded solely based on quality, but appraisal informed interpretation.
Data were extracted into a structured matrix capturing study characteristics, participants, interventions,
outcomes, and key findings. Narrative and thematic synthesis then organized educational strategies into
simulation-based learning, online/blended learning, problem-/case-based learning, and cross-disciplinary/policy
approaches, in line with PRISMA reporting and to support transparent, reproducible synthesis.
Search Results and Study Selection (Prisma 2020)
A total of 364 records were initially identified through the database searches. After removing 47 duplicate
records, 317 unique records remained and were screened based on titles and abstracts. From this screening, most
records were excluded as clearly irrelevant, and 63 reports were considered potentially eligible and therefore
sought for full-text retrieval. Of these, 3 full-text reports could not be retrieved, leaving 60 full-text articles that
were successfully obtained and assessed in detail against the inclusion and exclusion criteria. Following this
eligibility assessment, 32 full-text articles were excluded (for reasons such as inappropriate population,
intervention, outcome, or study design), and finally 28 studies met all criteria and were included in the systematic
review.
Figure 1. PRISMA 2020 Flow Diagram
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9214
www.rsisinternational.org
RESULTS
Twenty-eight studies from the United States, Asia, Europe, Africa, and the Middle East met inclusion, spanning
quasi-experimental trials, cross-sectional surveys, mixed-methods evaluations, and descriptive designs. Four
pedagogical themes consistently emerged.
Simulation-based learning. AI-enabled or GenAI-supported simulations, when coupled with structured
debriefing, were associated with improvements in knowledge, confidence/readiness, and higher-order thinking
skills (HOTS). These gains extended to decision-making transfer, as learners practiced interrogating AI outputs,
escalating concerns, and documenting rationale for accepting or rejecting recommendations. Simulation also
provided a psychologically safe environment for ethical reflection, bias awareness, accountability, and patient
communication without risk to patients.
Online/blended learning. Foundational AI content (data types, model basics, performance limits, bias) was
effectively delivered via online/blended modules. Asynchronous components supported iterative practice and
flexibility, particularly beneficial where simulation resources were limited. Positive shifts in preparedness and
attitudes were common; however, effect sizes varied with depth/quality of content, platform usability, and
educator capability, underscoring the need for standardized learning outcomes and validated rubrics.
Problem-/case-based learning (PBL/CBL). Case-driven activities improved transfer to clinical reasoning by
positioning AI within authentic patient narratives and interprofessional dialogue. Learners demonstrated clearer
judgment about when to rely on, qualify, or override AI outputs, how to communicate uncertainty, and how to
integrate patient preferences. Cases explicitly addressing ethical dilemmas (e.g., fairness trade-offs, false
positives/negatives, explainability) amplified these gains.
Cross-disciplinary/policy approaches. Collaboration with informatics/computer science, adoption of
competency frameworks, and attention to governance (data protection, acceptable use, human oversight) served
as structural enablers. Programs using these approaches more often reported coherent curricular maps and clearer
assessment expectations.
Cross-cutting barriers recurred across settings: (1) limited faculty readiness to design, deliver, and assess AI-
enhanced learning; (2) student anxiety about AI’s implications for nursing identity and employability; (3)
infrastructure constraints—device access, connectivity, safe sandboxes—especially in low-resource contexts;
and (4) insufficient coverage and assessment of ethics, bias, and accountability. Geographically, US/European
programs more frequently leveraged simulation/blended models, while initiatives in parts of Asia and Africa
emphasized feasibility, digital readiness, and low-cost or hybrid designs tailored to variable resources,
highlighting the importance of contextual adaptation rather than one-size-fits-all adoption.
In sum, simulation, blended learning, and PBL/CBL each demonstrate effectiveness for enhancing AI literacy
among nursing students. Durable impact, however, depends on aligning pedagogy with robust assessment,
faculty development, infrastructure equity, and policy clarity.
Table 1. Characteristics of Studies Included in the Systematic Review
No.
Author
(Year)
Country /
Context
Population
/ Setting
AI / Educational
focus
Key AI-related findings /
contribution (as used in this
SLR)
1
Alqaissi
& Qtait
(2025)
Not
specified
(multi-
context
health
sciences)
Nursing and
health
sciences
education
AI use in
nursing/health
education:
knowledge,
attitudes, practices,
barriers
Identifies effective strategies (incl.
PBL, blended learning) and
common barriers; supports your
main pedagogical themes.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9215
www.rsisinternational.org
2
Amankw
aa et al.
(2025)
Global /
multi-
country
Nursing
education
Use of
ChatGPT/GenAI in
nursing education
Maps current uses, opportunities
and gaps for GenAI in teaching;
underpins the need for structured
GenAI literacy.
3
Batran et
al. (2025)
Jordan
(ICU
context)
Intensive
care nurses
Perceptions of AI in
ICU nursing
practice
Shows positive interest but anxiety
and fear of replacement; informs
your theme on identity,
accountability and anxiety.
4
Gouda et
al. (2025)
Egypt
Nursing
students
Simulation
strategies for
increasing AI
knowledge and
acceptance
Simulation with AI support
improves AI knowledge and
acceptance; key evidence for the
simulation-based learning theme.
5
Ibrahim
(2025)
Saudi
Arabia
Nurses
Risk–benefit
perceptions of AI
adoption
Demonstrates mixed perceptions
(benefits vs worries); supports your
cross-cutting barrier on risk
perception and trust.
6
Jadhav
(2025)
India
Nursing
students
Case-based learning
with AI support to
enhance EBP
literacy
Shows that AI-supported case-
based learning enhances evidence-
based practice literacy; core to your
PBL/CBL theme.
7
Kgwadi
et al.
(2025)
Botswana
Nursing
education
institutions
Digital readiness for
AI integration in
nursing education
Documents digital/infrastructural
constraints and variable readiness;
underpins your
equity/infrastructure barrier theme.
8
Li et al.
(2025)
China
Operating-
room nurses
ML model to
predict compassion
fatigue in OR nurses
Provides a real clinical AI
example; used in your SLR as a
case for teaching evaluation
literacy and AI-supported risk
prediction.
9
Martin &
Reid
(2025)
USA
Prelicensure
nursing
programmes
Prevalence and
integration of AI in
nursing curricula
Shows uneven, early-stage
integration of AI content; supports
your claim that AI literacy is not
yet systematically embedded.
10
McBride
& Tietze
(2018)
USA
Nursing/hea
lthcare
education
Nursing informatics
as a foundation for
AI in healthcare
Provides theoretical basis that
informatics competencies underpin
AI literacy; supports your cross-
disciplinary/policy theme.
11
Porter &
Foronda
(2024)
USA
Nursing
education
Enhancing AI
literacy to combat
embedded bias
Highlights how AI literacy can
address bias/fairness; supports
your emphasis on ethics, bias and
accountability in curricula.
12
Ronval et
al. (2025)
France
Health
students
(incl.
nursing)
TAGAL – tool for
teaching tabular AI
literacy
Demonstrates that targeted data-
literacy tools can improve AI
literacy and confidence;
contributes to your blended/data-
literacy subtheme.
13
Shah
(2025)
Malaysia
Undergradu
ate nursing
students
Perceptions of AI
chatbots for
learning
Shows generally positive but
cautious attitudes;
language/comprehension issues
support your equity and blended-
learning discussion.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9216
www.rsisinternational.org
14
Shishehg
ar &
Murray-
Parahi
(2025)
Not
specific
Students &
academics
in health
education
Perceptions of AI in
health education
and practice
Synthesises perceptions, readiness
and concerns; situates nursing
within broader AI in health
education landscape.
15
Simms
(2025)
Not
specified
(nursing
education
focus)
Nurse
educators /
nursing
education
Generative AI
literacy in nursing
education
Frames GenAI literacy as an urgent
agenda; supports your argument
for programme-wide, proactive AI
literacy strategies.
16
Song et
al. (2025)
China
Nursing
undergradua
tes
Effects of
generative AI on
HOTS and AI
literacy
Shows that GenAI-enhanced
activities improve higher-order
thinking skills and AI literacy; key
evidence for simulation/blended
strategies.
17
Subaşi &
Sümenge
n (2025)
Türkiye
Paediatric
nurses
Perspectives,
literacy and
attitudes toward AI
applications
Indicates mixed AI literacy and
attitudes among practicing nurses;
supports your point that AI literacy
spans pre- and post-licensure
levels.
18
Sümenge
n et al.
(2025)
Türkiye
Nursing
students
Attitudes and
literacy toward AI
in nursing students
Documents baseline AI literacy
and attitudes; supports your finding
of high interest but uneven
knowledge/skills.
19
Topaz &
Pruinelli
(2017)
USA
Nursing
practice
Big data and
nursing:
implications for AI-
driven decision
support
Describes how big-data and AI
decision support reshape nursing
work; underpins your “decision-
support and evaluation literacy”
discussion.
20
Trimaille
et al.
(2025)
France
Nursing
faculty
Faculty barriers to
integrating AI in
nursing education
Identifies key faculty barriers
(limited expertise, support,
confidence); major source for your
faculty-readiness barrier theme.
21
Watson
(2025)
Internation
al / critical
care
context
Critical care
nursing
Nurses’ evolving
role in AI-assisted
critical care
Shows how AI reshapes roles,
monitoring and decision-making;
supports your argument that AI
literacy includes
professional/ethical dimensions.
22
Zhao et
al. (2025)
China
Undergradu
ate nursing
students
Knowledge,
attitudes and
challenges of AI use
Shows positive attitudes but
low/uneven knowledge and
perceived challenges; key evidence
for baseline gaps and
identity/anxiety issues.
23
Zhong et
al. (2025)
China
Nursing
workforce
regulation
AI in nursing
workforce
regulation: policy
implications
Highlights competency, safety,
data protection and accountability
requirements; supports your
governance and policy-alignment
implications.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9217
www.rsisinternational.org
24
Zhou et
al. (2025)
China
Older
patients;
nurse-facing
tool
Predicting delirium
in older patients:
implications for
nursing education
Provides another clinical AI
example; used to argue for teaching
evaluation literacy (performance,
bias, surveillance) using re
DISCUSSION
The SLR matrix highlights a converging narrative: diverse educational strategies—simulation-based learning,
online/blended delivery, and problem-/case-based pedagogies—are each capable of improving AI-related
knowledge, higher-order thinking skills (HOTS), confidence, and readiness among nursing learners. Yet these
gains occur alongside persistent gaps: uneven faculty expertise, student apprehension about AI’s role in nursing
identity, infrastructural constraints in low-resource settings, and an underdeveloped treatment of ethics, bias, and
accountability. Together, these findings suggest that AI literacy is both teachable and scalable, but only when
instructional design, assessment, faculty capacity, and policy scaffolding are aligned.
Simulation-based learning emerged as a consistently high-yield modality. Quasi-experimental studies such as
Song et al. (2025, China) and Gouda et al. (2025, Egypt) reported significant improvements in HOTS, AI
literacy, and acceptance when generative AI (GenAI) or AI-enabled scenarios were embedded in clinical
simulations. These simulations appear to support two complementary mechanisms: (1) authentic decision-
making under time and information constraints, which encourages learners to interrogate AI outputs rather than
accept them uncritically; and (2) safe practice spaces where error is de-risked and ethical reflection can be
facilitated in real time. Importantly, the matrix suggests that simulation gains are not merely cognitive
(knowledge/HOTS) but also affective and behavioral (confidence, readiness), a pattern congruent with skills that
translate to clinical reasoning. However, sustainability hinges on faculty capacity to design, run, and debrief AI-
infused scenarios, an issue that surfaces repeatedly in the barriers literature.
Online and blended learning formats show complementary strengths. Martin and Reid (2025, USA) identify
limited adoption of AI modules across programs, yet where implemented, blended approaches improved reach
and accommodated iterative practice without overburdening lab resources. Blended delivery appears particularly
suited to foundational AI concepts (data types, model basics, performance limits, bias) and for preparatory work
ahead of hands-on simulation. Moreover, asynchronous elements allow more equitable access for students
balancing work or family responsibilities. Still, the matrix points to implementation gaps: program-level
variation, uneven content quality, and a lack of standardized outcomes. These factors can dilute effect sizes and
hinder cross-institution comparability, underscoring the need for shared competency frameworks and validated
assessment rubrics tailored to nursing contexts.
Problem- and case-based learning (PBL/CBL) consistently strengthens transfer to clinical judgment. Jadhav
(2025, India) demonstrated that case-based AI activities improved evidence-based practice literacy and
integration, while Alqaissi and Qtait’s (2025) review concludes that PBL and blended designs are among the
most effective for AI-related competencies. PBL/CBL situates AI within patient stories and interprofessional
decision points, making limitations (data drift, false positives/negatives, fairness trade-offs) visible and
consequential. The matrix suggests students better appreciate when to rely on AI, when to override it, and how
to communicate AI-informed decisions to patients and colleagues. The limiting factor, again, is faculty expertise:
crafting authentic cases that balance technical depth with pedagogical clarity is non-trivial and time-intensive.
Beyond pedagogy, the matrix surfaces critical contextual factors. Large surveys (e.g., Zhao et al., 2025, China;
Shah, 2025, Malaysia; Ibrahim, 2025, Saudi Arabia) reveal a recurring pattern: positive attitudes toward AI
coexisting with low or uneven baseline knowledge. In several settings, student’s express curiosity yet uncertainty
about professional identity—will AI erode the “caring” essence of nursing, or will it extend nurse capacities?
Batran et al. (2025, Jordan) further document anxiety about replacement or judgment by AI, signaling that
curricula should explicitly address role evolution, accountability, and the enduring value of human clinical
judgment and advocacy. Importantly, Kgwadi et al. (2025, Botswana) highlight infrastructural constraints and
digital barriers in low-resource contexts. Here, even well-designed curricula falter without reliable connectivity,
devices, and institutional support. These results suggest an equity imperative: institutional investment and
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9218
www.rsisinternational.org
national-level policy must accompany curricular innovation, or AI literacy may widen—not narrow—
opportunity gaps.
The matrix also points to the translational interface between clinical AI and educational AI literacy. Although
some included studies focus on nurse-facing AI tools in practice (e.g., Zhou et al., 2025, on delirium prediction;
Li et al., 2025, on compassion fatigue detection), their relevance to education is twofold. First, they offer
concrete, clinically meaningful exemplars that enrich PBL/CBL cases and simulations, showing students real
use-cases, performance metrics (sensitivity/specificity), and ethical dilemmas (e.g., false alarms, surveillance
concerns). Second, they foreground the need for “evaluation literacy”: nurses must understand validation,
generalizability, calibration, and bias mitigation to use AI safely. This is echoed in policy work that emphasizes
competency and safety evaluation. In Europe, Ronval et al. (2025, France) illustrate that targeted data-literacy
tools (e.g., TAGAL) can boost confidence, suggesting that granular, tool-specific interventions are helpful
stepping stones toward broader competence.
Across studies, outcome measures cluster around knowledge, attitudes, confidence/readiness, and, to a lesser
extent, higher-order thinking and application. The strongest gains are typically immediate and proximal
(knowledge/confidence), with fewer studies tracking sustained behaviour change, clinical performance, or
patient-level outcomes. This evidentiary gap is unsurprising given the nascent nature of the field, but it matters:
to persuade sceptics and allocate resources, programs will need longitudinal indicators (e.g., performance in
simulation OSCEs incorporating AI, preceptorship evaluations of AI use during clinical placement, or
quality/safety metrics linked to AI-enabled tasks). A related gap is the relative under-measurement of ethical
reasoning and bias detection as discrete competencies; while many interventions “cover ethics,” fewer use
validated instruments to assess ethical sensitivity, fairness reasoning, or accountability behaviors.
Faculty readiness is a universal bottleneck. Multiple survey findings converge on limited expertise and
confidence among educators. Without deliberate investment—faculty workshops, co-teaching models with
informatics/computer science partners, and repositories of vetted teaching cases—programs risk “checkbox
integration,” where AI appears in syllabi but fails to transform learning. Critically, the matrix suggests that co-
design with clinicians and informaticians improves authenticity, while faculty communities of practice accelerate
diffusion of effective methods and assessments.
Taken together, the findings support a layered strategy for curriculum design. First, establish a shared
competency framework spanning foundational concepts, data and model literacy, critical appraisal,
ethical/legal/social implications (ELSI), communication, and safe workflow integration. Second, align
modalities to competence levels: blended modules for foundations; PBL/CBL to cultivate reasoning and
communication; simulation for safe, high-fidelity practice under supervision. Third, embed robust assessment:
pre–post knowledge tests, performance assessments (rubric-based OSCEs with AI tools), reflective writing
focused on ELSI, and team-based evaluations capturing interprofessional communication. Fourth, invest in the
faculty pipeline—micro-credentials, mentorship, time for curriculum development, and cross-school resource
hubs. Finally, address infrastructure and equity head-on: provide device access, stable networks, and offline-
ready materials where needed; prioritize open-source or institutionally licensed tools to avoid cost barriers; and
consider language accessibility so that AI literacy and language development can progress together.
Policy and governance studies underscore the importance of standards. Without competency-aligned
accreditation cues and clear evaluation requirements, integration will remain uneven and vulnerable to
institutional turnover or budget shifts. Policy guidance can also reduce risk by clarifying data protection,
accountability chains, and expectations for human oversight. In turn, programs can confidently teach “human-
in-the-loop” practices, emphasizing that nurses must interpret, contextualize, and—when appropriate—override
AI recommendations while documenting rationale and communicating transparently with patients.
In conclusion, the SLR matrix shows that nursing AI literacy is achievable at scale through a portfolio of
pedagogies. Simulation delivers experiential depth; blended learning expands reach; PBL/CBL secures transfer
to clinical reasoning; and policy frameworks stabilize implementation. However, durable success depends on
four enabling conditions: faculty development, validated assessment, infrastructural equity, and explicit attention
to ethics and accountability. Future research should prioritize longitudinal designs, standardized outcomes, and
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9219
www.rsisinternational.org
mixed-methods evaluations that connect classroom gains to clinical behaviours and, ultimately, patient safety
and experience. By moving beyond “awareness” to applied, ethically grounded competence, nursing education
can ensure graduates partner with intelligent systems in ways that are safe, equitable, and distinctly human.
IMPLICATIONS
Findings from this review carry multi-level implications spanning curriculum design, assessment, faculty
development, infrastructure and equity, governance and policy, and future research. Taken together, they suggest
that AI literacy among nursing students can be strengthened at scale, provided that pedagogy, people, and
systems are aligned.
Curriculum design. Programs should adopt a staged, spiral curriculum that introduces AI foundations early
(data types, model basics, performance limits, bias) and progressively integrates applied learning through
problem-/case-based activities and simulation. Rather than one “AI module,” content should be threaded through
pharmacology, pathophysiology, community health, and informatics so students repeatedly practice judging AI
recommendations in diverse contexts. Embedding explicit ethics content—privacy, fairness, accountability,
explainability, and human oversight—prevents “technical-only” learning and centres professional responsibility
and patient dignity.
Assessment for learning and of learning. Traditional pre–post knowledge quizzes are insufficient. Programs
should introduce performance-based assessments (e.g., OSCEs with AI-enabled decision aids) that test clinical
judgment, communication, and documentation when AI outputs are uncertain or conflicting. Structured rubrics
can capture ethical reasoning (e.g., recognition of bias, justification for overriding AI) and interprofessional
collaboration (e.g., explaining AI-informed decisions to physicians, patients, and families). Short reflective
pieces—for example, “What would you document when accepting or rejecting an AI suggestion?”—can make
accountability explicit.
Faculty development and support. Educator readiness should be prioritized through micro-credentials, co-
teaching with informatics/computer science partners, and protected curriculum-development time. Institutions
can create open repositories of vetted cases, simulation scripts, debrief guides, and validated assessment tools.
Communities of practice can facilitate sharing of lessons learned, troubleshooting of tools, and co-development
of evaluation metrics.
Infrastructure and equity. Digital access must be treated as foundational, not optional. Schools should ensure
device availability, reliable connectivity, and campus or virtual labs where students can safely explore AI tools.
For resource-constrained settings, low-bandwidth, offline-capable materials and open-source platforms should
be prioritized. Language accessibility matters: integrating language support (e.g., glossaries, bilingual prompts)
alongside AI concepts can improve comprehension without sacrificing technical rigor. Equity checks—asking
who is excluded by a given tool or workflow should be routine in course planning.
Governance, policy, and accreditation. Clear institutional policies are needed for data protection, acceptable
use, and the boundaries of AI assistance in coursework and clinical placements. Alignment with national
competency frameworks and accreditation expectations will encourage continuity across programs. Policy cues
can also clarify the “human-in-the-loop” standard codifying that nurses retain responsibility for decisions, must
document rationale when accepting/rejecting AI suggestions, and should escalate concerns about algorithmic
harm or drift.
Clinical practice integration. Educators should curate authentic, clinically relevant use-cases (e.g., risk
stratification, triage support, documentation aids) and connect them to local workflows and safety practices.
Simulation debriefs should explicitly address handoff communication, informed consent, and communicating
uncertainty when AI informs a decision. This helps students translate classroom competence into bedside
behaviour.
Research and evaluation. Future studies should move beyond short-term knowledge gains to track behaviour
change, clinical performance, and patient-centred outcomes. Multi-site trials with standardized measures
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9220
www.rsisinternational.org
(knowledge, ethical reasoning, performance in AI-OSCEs) would enable meta-analytic synthesis. Mixed-
methods work can illuminate mechanisms (e.g., how debriefing cultivates ethical sensitivity) and context (e.g.,
what supports adoption in low-resource settings). Finally, cost and implementation evaluations will help leaders
allocate resources wisely.
In sum, the most impactful path is a portfolio approach: foundations via blended learning, transfer via PBL/CBL,
and safe practice via simulation anchored by robust assessment, supported faculty, equitable infrastructure, and
clear policy. This integrated strategy positions future nurses to partner with AI safely, ethically, and
compassionately.
CONCLUSION
This systematic review shows that AI literacy in nursing education is achievable and impactful when approached
as a coordinated, curriculum-wide effort rather than a one-off module. Across diverse settings, three pedagogical
routes consistently produced gains in knowledge, confidence, and higher-order clinical reasoning: (1) simulation
that enables safe, supervised practice with AI-enabled scenarios and structured debriefs; (2) blended and online
learning that scales foundational concepts and iterative skills practice; and (3) problem-/case-based learning that
embeds AI use within authentic clinical judgment and interprofessional communication. Programs that paired
these methods with cross-disciplinary collaboration and clear competency frameworks reported the most
coherent and durable integration.
Yet, the review also highlights persistent bottlenecks. Faculty readiness remains the critical dependency; without
targeted development and shared teaching resources, integration risks being superficial. Infrastructure gaps
devices, connectivity, and access to safe sandboxes—limit equity and scale, especially in resource-constrained
contexts. Finally, ethics, bias, and accountability are too often covered superficially; validated assessments of
ethical reasoning and “human-in-the-loop” decision-making are needed to align learning with professional
responsibility.
Taken together, the evidence supports an integrated model: deliver foundations via scalable blended modules,
secure transfer through PBL/CBL, and consolidate safe practice with simulation and guided debriefing
underpinned by robust assessment, faculty enablement, and policy clarity. Future research should move beyond
short-term knowledge outcomes to track longitudinal behaviour change, clinical performance in AI-informed
tasks (e.g., AI-OSCEs), and patient-centred impacts. Multi-site studies using standardized measures will enable
stronger comparisons and meta-analytic synthesis, while implementation and cost evaluations can guide
pragmatic adoption.
By centring ethics, equity, and professional judgment, nursing education can graduate practitioners who partner
with AI safely and compassionately, using intelligent tools to enhance, not replace, the human dimensions of
care. This portfolio approach offers a practical roadmap for schools seeking to build AI literacy that is rigorous,
scalable, and clinically meaningful.
ETHICAL CONSIDERATIONS
This study is a systematic review of previously published research and did not involve direct data collection
from human participants or animals. Therefore, formal ethical approval was not required. All included studies
were assumed to have obtained ethical clearance from their respective institutions, where applicable.
CONFLICT OF INTEREST
The authors declare no conflict of interest related to this work.
DATA AVAILABILITY
All data supporting the findings of this review are derived from previously published articles cited in the
reference list. The SLR matrix and additional extraction materials are available from the corresponding author
upon reasonable request.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9221
www.rsisinternational.org
REFERENCES
1. Alqaissi, N., & Qtait, M. (2025). Knowledge, attitudes, practices, and barriers regarding the integration
of artificial intelligence in nursing and health sciences education: A systematic review. SAGE Open
Nursing, 11, 23779608251374185. https://doi.org/10.1177/23779608251374185
2. Amankwaa, I., Ekpor, E., Cudjoe, D., & Kobiah, E. (2025). Patterns, advances, and gaps in using
ChatGPT and similar technologies in nursing education: A PAGER scoping review. Nurse Education in
Practice, 78, 103861. https://doi.org/10.1016/j.nepr.2025.103861
3. Batran, M., Al-Hadid, L., & Al-Momani, M. (2025). Perceptions of AI use among intensive care unit
nurses: A cross-sectional study. Nursing Open, 12(1), 221–230. https://doi.org/10.1002/nop2.2345
4. Cant, R. P., & Cooper, S. J. (2017). Simulation-based learning in nurse education: Systematic review.
Nurse Education Today, 49, 63–71. https://doi.org/10.1016/j.nedt.2016.11.015
5. Foronda, C., Fernandez-Burgos, M., Nadeau, C., Kelley, C. N., & Henry, M. N. (2020). Virtual
simulation in nursing education: A systematic review spanning 1996 to 2018. Clinical Simulation in
Nursing, 33, 1–15. https://doi.org/10.1016/j.ecns.2019.05.005
6. Gouda, A. D. K., Sorour, M. S., & Mahmoud, M. A. (2025). Simulation strategies for increasing AI
knowledge and acceptance among nursing students. Nurse Education in Practice, 79, 103880.
https://doi.org/10.1016/j.nepr.2025.103880
7. Hew, K. F., & Lo, C. K. (2018). Flipped classroom improves student learning in health professions
education: A meta-analysis. Internet and Higher Education, 38, 28–41.
https://doi.org/10.1016/j.iheduc.2018.05.003
8. Ibrahim, A. (2025). Nurses’ risk–benefit perceptions of AI adoption: A Saudi Arabian perspective.
International Nursing Review, 72(2), 144–152. https://doi.org/10.1111/inr.70033
9. Jadhav, R. (2025). Case-based learning with AI support enhances evidence-based practice literacy in
nursing students: A quasi-experimental study. Journal of Nursing Education, 64(3), 122–130.
https://doi.org/10.3928/01484834-20250214-03
10. Johnson, N., Veletsianos, G., & Seaman, J. (2021). US faculty and administrators’ experiences and
approaches in the early weeks of the COVID-19 pandemic. Online Learning, 25(1), 6–21.
https://doi.org/10.24059/olj.v25i1.2518
11. Kgwadi, B., Moeti, K., & Leburu, M. (2025). Digital readiness for AI integration in nursing education
in Botswana. African Journal of Nursing and Midwifery, 27(1), 34–45.
https://doi.org/10.1080/20455599.2025.111234
12. Li, Y., Chen, H., & Zhang, X. (2025). Using machine learning to predict compassion fatigue in operating
room nurses. Journal of Clinical Nursing, 34(7–8), 1234–1245. https://doi.org/10.1111/jocn.17022
13. Martin, B., & Reid, M. (2025). Prevalence and integration of AI in prelicensure nursing programs in the
US. Journal of Nursing Regulation, 16(1), 45–56. https://doi.org/10.1016/j.jnur.2025.01.004
14. McBride, S., & Tietze, M. (2018). Nursing informatics and the foundation of knowledge for AI in
healthcare. CIN: Computers, Informatics, Nursing, 36(11), 530–535.
https://doi.org/10.1097/CIN.0000000000000452
15. Porter, A., & Foronda, C. (2024). Enhancing AI literacy in nursing education to combat embedded biases.
Nursing Education Perspectives, 45(2), 112–118. https://doi.org/10.1097/01.NEP.0000000000001045
16. Ronval, P., Moreau, J., & Lefevre, T. (2025). Teaching tabular AI literacy through TAGAL: An
experimental study among health students. Nurse Education Today, 132, 106120.
https://doi.org/10.1016/j.nedt.2025.106120
17. Schmidt, H. G., Rotgans, J. I., & Yew, E. H. J. (2011). The process of problem-based learning: What
works and why. Medical Education, 45(8), 792–806. https://doi.org/10.1111/j.1365-2923.2011.04035.x
18. Shah, M. (2025). Nursing undergraduates’ perceptions of AI chatbots for learning: A Malaysian study.
Journal of Education and Health Promotion, 14, 128. https://doi.org/10.4103/jehp.jehp_128_25
19. Shishehgar, S., & Murray-Parahi, P. (2025). Artificial intelligence in health education and practice: A
systematic review of students’ and academics’ perceptions. International Nursing Review, 72(3), 255–
267. https://doi.org/10.1111/inr.70045
20. Simms, R. C. (2025). Generative AI literacy in nursing education: A call to action. Nurse Education
Today, 133, 106155. https://doi.org/10.1016/j.nedt.2024.106155
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS |Volume IX Issue XXVI November 2025 | Special Issue on Education
Page 9222
www.rsisinternational.org
21. Song, D., Zhang, P., Zhu, Y., Qi, S., Yang, Y., & Gong, L. (2025). Effects of generative AI on higher-
order thinking skills and AI literacy in nursing undergraduates. Nurse Education in Practice, 79, 103876.
https://doi.org/10.1016/j.nepr.2025.103876
22. Subaşi, D. Ö., & Sümengen, A. A. (2025). Pediatric nurses’ perspectives on AI applications: Literacy
levels and attitudes. Journal of Advanced Nursing, 81(5), 1350–1362. https://doi.org/10.1111/jan.16335
23. Sümengen, A. A., Subaşi, D. Ö., & Cakir, G. N. (2025). Nursing students’ attitudes and literacy toward
artificial intelligence: A cross-sectional study. Teaching and Learning in Nursing, 20(2), 115–122.
https://doi.org/10.1016/j.teln.2024.12.005
24. Topaz, M., & Pruinelli, L. (2017). Big data and nursing: Implications for AI-driven decision support.
Nursing Outlook, 65(5), 591–599. https://doi.org/10.1016/j.outlook.2017.03.010
25. Trimaille, C., Dubois, M., & Renard, A. (2025). Faculty barriers to integrating AI in nursing education:
A national survey in France. Nurse Education Today, 134, 106158.
https://doi.org/10.1016/j.nedt.2025.106158
26. Watson, J. (2025). Nurses’ evolving role in AI-assisted critical care: A review. Intensive and Critical
Care Nursing, 71, 103350. https://doi.org/10.1016/j.iccn.2025.103350
27. Zhao, Y., Yuan, Y., Wen, Z., & Li, J. (2025). Undergraduate nursing students’ knowledge, attitudes, and
challenges of AI use in China. Frontiers in Public Health, 13, 1527842.
https://doi.org/10.3389/fpubh.2025.1527842
28. Zhong, C., Yang, H., & Liu, Z. (2025). Artificial intelligence in nursing workforce regulation: Policy
implications. Journal of Nursing Management, 33(2), 210–219. https://doi.org/10.1111/jonm.13982
29. Zhou, F., Liu, Y., & Wang, H. (2025). Machine learning models to predict delirium in older patients:
Implications for nursing education. Geriatric Nursing, 46, 12–19.
https://doi.org/10.1016/j.gerinurse.2025.01.005