Redefining pedagogy with artificial intelligence: How nursing students are shaping the future of learning

Synthesis: Ghimire and Qiu report an exploratory, descriptive qualitative study of 17 final-year undergraduate nursing students at one Australian metropolitan university, 10 domestic and 7 international, interviewed for approximately 60 minutes each and analyzed with Braun and Clarke's six-phase thematic analysis. Six themes describe AI as a personalized learning accelerator, a clinical confidence builder, a time management multiplier, an equalizer across the digital divide, a student-driven answer to institutional inertia, and an object of critical ethical awareness. The finding the authors foreground is the "open secret": students use AI openly among themselves while the institution discourages it, which they read as a problem of trust and transparency in the student-educator relationship rather than a problem of cheating.

Key Findings

1. AI as a personalized learning accelerator. Students uploaded lecture notes and textbook chapters to generate quizzes, flashcards, and predicted exam questions, and used immediate feedback to drill weak areas such as pharmacology calculations. The authors frame this as bridging the gap between theory and practice rather than shortcutting it.
2. AI as a clinical confidence builder. Virtual patient simulations let students rehearse emergencies, culturally sensitive communication, and other high-stakes encounters without risk to real patients. One domestic student described feeling calm at a first real emergency because the simulation had already absorbed the mistakes.
3. AI as a time management and efficiency multiplier. Mature students with children and international students working part time used AI to compress literature searching, outlining, and grammar checking. Time saved was described as time returned to family or to paid work.
4. The digital divide is about learning experience, not only hardware. Text to speech, summaries, and visual aids made study feasible for students with learning disabilities and for students whose first language was not English. The authors argue the divide in nursing education now runs through access to effective learning, not just access to devices.
5. The "open secret." Students shared tools and tips among themselves while treating AI as unmentionable in class, where lecturers framed it as cheating or as a way of avoiding critical thinking. Students described the gap as hypocrisy rather than as rebellion.
6. Ethical awareness was already present. Students raised algorithmic bias, patient privacy when clinical information is entered into tools, and accountability when AI output is wrong, and they placed human oversight ahead of tool capability.

An exploratory qualitative design at one metropolitan school

The study took place in a metropolitan tertiary institution in Melbourne, Australia, and recruited a purposive sample from its undergraduate Bachelor of Nursing program. All 17 participants were final-year students: 10 domestic and 7 international, 12 female and 5 male, and 14 of the 17 said nursing was their first choice of career. Final-year students were targeted because clinical placements give them a view of how AI might be used in the profession, not just in coursework.

Data came from 17 individual, semi-structured interviews of approximately 60 minutes, conducted face to face by the first author in a location convenient to participants, audio recorded with consent, and transcribed verbatim by the principal investigator. No prior relationship existed between researchers and participants, and the study follows the COREQ reporting guidelines. The interview guide asked what tools students used, how use affected learning, what benefits and challenges they saw, how AI compared with traditional teaching, what guidance they wanted from instructors, and what ethical concerns they held.

Analysis followed Braun and Clarke's six-phase thematic analysis. Two researchers coded independently, the codebook was refined through comparison and consensus meetings, and the study aimed at code and thematic saturation rather than data saturation. Trustworthiness rested on two phases of member checking: the interviewer summarized key points during each interview for immediate confirmation, and after analysis a summary of the themes and supporting quotes went by email to all 17 participants, who had two weeks to respond. Peer debriefing by an external expert and reflexive journals kept by a team of nursing academics completed the credibility work, and the Monash University Human Research Ethics Committee approved the study under approval number 46091.

What the six themes name

The first theme, "AI as a personalized learning accelerator: bridging the gap between theory and practice," covers tailored study plans, practice quizzes, and simulations, and the adaptation to unfamiliar teaching styles and assessment formats. It is where the international students' experience first becomes visible: AI translation for complex medical terms, culturally relevant case studies, and a simulation that let one student rehearse patient interactions across cultural contexts. The second theme, "AI as a clinical confidence builder: navigating the complexities of real-world scenarios," is about practicing complex scenarios without fear of harming real patients, including culturally sensitive communication and nonverbal cues that differ between home and host country.

The third theme, "AI as a time management and efficiency multiplier: balancing academic demands and personal life," is explicit about who benefits. A mature student described the arithmetic of care: "Being a mature student with two young children, time is my most precious commodity" (SN1, Domestic Student). An international student described using AI during a late shift to get an outline and a grammar check on an assignment. The fourth theme, "Bridging the Digital Divide: AI as an Equalizer for Access to Quality Education," reframes access as a question of whether the learning format fits the learner, and it is here that the strongest claim for inclusion appears: "It was like finally having a key to unlock my learning potential" (SN14, International Student).

The fifth theme, "overcoming institutional inertia: student-driven demand for ai integration in nursing curricula," carries the paper's central tension. "The biggest irony is that everyone knows we're using AI" (SN9, Domestic Student), the same student said, before reaching for the analogy the authors return to in the discussion: "It's like they are forcing us to learn how to drive a car but are not letting us use the GPS" (SN9, Domestic Student). The sixth theme, "beyond the hype: critical evaluation and ethical awareness of ai in nursing practice," shows students arriving at critique without being taught it: "It made me realize that AI isn't objective; it's trained on data created by humans and humans have biases" (SN4, Domestic Student). Privacy and accountability followed the same pattern, with students asking who is responsible when a tool is wrong and insisting that human oversight is not optional.

The open secret and the trust problem

The authors treat the "open secret" as the study's key contribution, and they are careful about what it means. Students did not describe a scheme to cheat. They were describing a workaround: a set of practices that helped them understand material, prepare for placements, and manage competing demands, conducted in a culture where disclosure carried risk. The paper connects this to work on student-teacher trust, where students are expected to declare AI use while assessment criteria and institutional positions remain opaque, a one-way transparency that produces surveillance rather than dialogue.

That reading changes what a response should look like. If the problem is a gap between student practice and institutional guidance, then tightening detection or repeating prohibitions addresses the symptom. The paper's recommendations point the other way: AI literacy hubs that bring students, educators, computer scientists, ethicists, and clinicians together; clinical simulation ecosystems that go beyond isolated tools; ethical AI use badging for students and educators; and policy co-created with student representatives rather than handed down. AI literacy here means more than operational skill. It means understanding how the systems work, where they fail, and what Privacy and accountability require when patient information is involved.

The convergence and divergence the authors report matter for how such guidance is written. Domestic and international students agreed that AI made study more efficient and agreed in their frustration with institutional resistance. They diverged on purpose: international students emphasized cultural and linguistic adaptation, domestic students emphasized learning style needs and time pressure from family or work. A single blanket policy addresses neither group well, and the study's finding that students are already shaping their own practice, an exercise of Learner Agency, is the reason to ask them.

What this means for practice

• Instructors. State permitted and prohibited uses in the course outline and say why. In this study, silence functioned as prohibition, and students responded by hiding use rather than stopping it.
• Curriculum designers. Build AI literacy into the program, including bias, privacy, and accountability, instead of leaving it to informal peer networks that already circulate tips.
• Simulation leads. Prioritize virtual patient scenarios for rare emergencies and for culturally sensitive communication, the two areas where students reported the largest confidence gains.
• Administrators. Co-create AI policy with student representatives and publish the reasoning, so that declaration requirements do not read as one-way surveillance.
• Assessment designers. Design tasks that make AI-assisted process visible, such as outlines and revision trails, rather than relying on bans that students experience as unenforceable.

Limitations

• The study is set in a single metropolitan tertiary institution in Melbourne, in one Bachelor of Nursing program. The authors state that the findings are not generalizable to all nursing students, and this page does not treat them as a description of nursing students generally.
• The sample is 17 final-year students who had used AI or wanted to explore it, so students who avoid AI, or who are earlier in the program, are absent from the account.
• Data are self-reported interviews only, with no observation or logging of actual tool use. Given the "open secret" finding, disclosure may be understated, and recall and social desirability bias are live risks the authors acknowledge.
• The paper reports member checking with all 17 participants but not how many responded, and the analysis relied on consensus between two coders rather than a reported reliability statistic.
• The study is a snapshot at one point in time, and the authors call for longitudinal work on how student AI practice carries into professional practice.

Citation

Ghimire, A., & Qiu, Y. (2025). Redefining pedagogy with artificial intelligence: How nursing students are shaping the future of learning. Nurse Education in Practice, 84, 104330.