Generative Artificial Intelligence in Laboratory-Based Health Sciences Education: A Scoping Review of an Emerging Evidence Base
This scoping review of nine recent studies reveals that while generative AI in laboratory-based health sciences education primarily enhances efficiency and supports individual inquiry and acquisition activities, its integration is currently limited by a small evidence base, risks of factual inaccuracy, and a notable lack of collaborative applications, underscoring the need for further comparative research and expert oversight.
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
Imagine the world of health science as a giant, high-stakes kitchen where the chefs don't just cook meals; they cook up the data that tells doctors how to save lives. In this kitchen, there are special experts called medical laboratory scientists. They are the ones who take blood, tissue, and other samples, run them through complex machines, and interpret the results. If they get the numbers wrong, the doctor might prescribe the wrong medicine. It's a job that requires a mix of hands-on skill, like handling tiny test tubes, and brainy skill, like spotting a weird pattern in a chart.
Recently, a new kind of "super-intelligent assistant" has crashed into this kitchen. It's called Generative Artificial Intelligence (AI). Think of it as a robot chef that can write recipes, explain cooking techniques, and even draft entire menus in seconds. It's based on Large Language Models, which are basically computers that have read almost everything on the internet and learned to predict what words should come next. The big question everyone is asking is: Can this robot chef actually help train the next generation of real human chefs, or is it just going to serve up delicious-looking but poisonous meals? We know it's great for writing essays or coding, but in a field where a tiny mistake can be dangerous, we need to know if it's safe to let it into the classroom.
This paper is a detective story that went looking for answers. The author, Kiran Zahid, didn't just guess; they went on a digital scavenger hunt to find every single study published between 2024 and 2026 that talked about using this AI robot in laboratory science education. They looked through three huge libraries of research (PubMed, Scopus, and ERIC) and found exactly nine studies that fit the rules. It's a tiny pile of evidence compared to the mountain of research on AI in nursing or general medicine, but it's the first time anyone has tried to map this specific corner of the map.
Here is what the detective found: The evidence is very new and very small. Out of the nine studies, most were about Pathology (the study of diseases) or general Biomedical Science. Surprisingly, the actual "Medical Laboratory Science" programs—the ones that train the people who run the labs—were barely mentioned, with only one study focusing on them. Also, the super-advanced stuff like molecular diagnostics (looking at DNA and genes) was almost completely ignored.
The studies showed that teachers are mostly using the AI to do their homework for them: writing lecture outlines, making up quiz questions, and drafting course materials. It's like the robot is helping the teacher write the menu before the class even starts. When it comes to students, the AI is mostly used for "Acquisition" (reading and learning facts) or "Inquiry" (asking questions). But here is the weird part: nobody is using it for "Discussion" or "Collaboration." Even though real lab work is all about teams talking to each other, the AI is currently being used to help students work alone.
The biggest discovery is a split personality in the findings. On one side, everyone agrees the AI is a fantastic time-saver. It helps teachers draft content faster and helps students get unstuck when they have writer's block. On the other side, everyone agrees the AI is a terrible liar. The most common problem reported was "hallucination," where the AI confidently makes up facts, invents fake references, or gives wrong numbers. In a lab, if the AI says a chemical level is safe when it's actually dangerous, that's a disaster. The studies found that the AI is especially bad at the specific, technical details of lab work, like knowing the exact reference ranges for blood tests or interpreting complex images.
Because of this, the studies suggest a clever workaround. Instead of trying to stop students from using the AI, three different groups of researchers independently came up with the same idea: use the AI's mistakes as a teaching tool. They gave students AI-generated essays or case studies that were full of subtle errors and asked the students to find and fix them. It's like a "spot the difference" game, but the differences are life-or-death medical errors. This turns the robot's biggest weakness—its tendency to lie—into its biggest strength as a teacher.
However, the paper is very clear about what it doesn't know. The author didn't find any proof that using this AI actually makes students smarter or better at their jobs. There are no big studies comparing students who use AI to those who don't. The evidence is mostly descriptive, meaning it tells us what people are trying to do, not whether it works. The paper concludes that while the technology is here to stay, we can't trust it blindly. The most important lesson for now is that students need to learn how to be detectives, checking the robot's work against the truth, because in the lab, trusting a confident-sounding robot without a human expert double-checking it is a recipe for trouble.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.