Content, implementation, and evaluation of artificial intelligence literacy education for nursing students : A scoping review
This scoping review of 14 sources reveals that AI literacy education for nursing students is an emerging field shifting from technical application to responsible, critical, and context-sensitive judgment, though it currently relies heavily on short-term assessments rather than performance-based evaluations in real-world nursing contexts.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Hospitals are changing. For decades, the tools of nursing have been stethoscopes, charts, and the steady hands of caregivers. Now, a new kind of tool has arrived: artificial intelligence. These are computer systems that can read, write, and solve problems, often mimicking human conversation. In the real world, they help doctors predict which patients might get sick and manage vast amounts of medical records. But for the students training to become nurses today, these tools present a unique challenge. It is not enough to simply know how to turn the machine on. A nurse must understand when to trust the machine, when to question its answers, and how to use it without losing the human connection that defines their profession. This new skill set is called AI literacy. It is the ability to use these technologies wisely, to spot their mistakes, and to keep patient safety at the center of every decision.
A team of researchers set out to see how nursing schools are teaching this vital skill. They did not look for a single perfect method, but rather cast a wide net to find any study, report, or curriculum that tried to teach nursing students how to work with artificial intelligence. They searched through thousands of records from databases around the world, looking for work published between early 2018 and mid-2026. After a careful review, they narrowed their focus to fourteen specific sources. These included experiments, classroom reports, and curriculum designs from countries like the United States, China, Australia, and Turkey. The students involved ranged from those in their first year of college to those earning advanced doctoral degrees.
What the researchers found was a field that is still very young but moving in a clear direction. The education is no longer just about showing students how to type a question into a computer. The focus has shifted toward something much deeper: responsible use and critical thinking. In the classrooms they studied, the most common lesson was not on the technical parts of the software, but on how to ask the right questions, known as "prompting," and how to check the answers the computer gives. Students were taught to treat the computer like a helpful but fallible assistant. They learned to verify facts, to look for hidden biases, and to ensure that any advice generated by the machine fit the specific needs of a real patient.
The way this teaching happens is often practical and hands-on. In many of the programs, students were given a task, such as writing a care plan or analyzing a case study, and then asked to use an AI tool to help them. The instructors did not just let them use the tool freely; they guided them through a process of generating an answer, checking it for errors, and then refining it. This cycle of "generate, verify, and revise" was the most common pattern. Students practiced using these tools to draft patient education materials or to organize complex medical information. However, the researchers noted that the most difficult part for students was not using the tool, but knowing when to stop relying on it. When the situations became highly complex or involved deep emotional needs, students often struggled to determine if the computer's suggestion was safe or appropriate.
The methods used to test whether students were learning were quite varied. Some schools used standard surveys where students rated their own confidence or knowledge. Others used grading rubrics to evaluate the quality of a care plan a student wrote with AI help. The results generally showed that students felt more comfortable with the technology and knew more about how it worked. They became better at spotting obvious errors in the computer's writing. Yet, the researchers pointed out a gap in the evidence. While students could pass a test or write a better report in a classroom setting, there is little proof yet that they can carry these skills into a busy, real-world hospital. We do not know if a student who learns to check an AI's work in a quiet classroom will do the same when a patient is in distress and time is short.
The studies also revealed that most of this education happens in short bursts, often within a single class or a few weeks of a semester. Very few programs have built a long-term plan that weaves AI literacy through all four years of a nursing degree. The researchers found that while some schools have started to create frameworks for this kind of long-term learning, they have not yet tested them with large groups of students. This means that the current landscape is a collection of promising experiments rather than a unified system. The teaching is happening, but it is scattered, and the ways of measuring success are not yet consistent.
Ultimately, the picture that emerges is one of cautious progress. Nursing schools are beginning to realize that AI literacy is not just a technical add-on, but a core part of professional judgment. The goal is to produce nurses who can use these powerful tools to enhance their care without letting the tools replace their own critical thinking. The researchers suggest that for this to work, education must move beyond short lessons and become a steady part of the entire nursing curriculum. Future programs need to focus less on how to use the software and more on how to judge its output in the messy, unpredictable reality of patient care. Until then, the field remains in its early stages, building a foundation for a future where human nurses and intelligent machines work together, with the human always holding the final responsibility for the patient's well-being.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.