Actionable companion · Instructional improvement
From evidence to district action: Can I Do Better Than AI: A Comparative Analysis of a Medical Student Essay Created With and Without Generative AI
AI-assisted editorial implementation guide. The district actions are proposed applications, not findings or recommendations attributed to the original authors. No additional peer review is asserted.
Evidence and context
Health-professions education context: test transfer to K–12 rather than assuming it.
Original authors’ abstract, reproduced under the linked source license:
Background With the increasing integration of generative artificial intelligence (AI) tools in education, questions have emerged about their effectiveness in academic tasks such as reviewing literature and writing essays. This paper evaluates whether generative AI could outperform a third-year medical student writing a literature review. Methods A third-year medical student wrote a 3000-word literature review on the prevention of breast cancer following University guidelines. Three prompting methods were used to generate essays with the same title and guidance using ‘ChatGPT-4o’ (OpenAI, San Francisco, CA, USA). Each essay was then graded alongside the student’s essay by an AI chatbot using the marking scheme provided. The highest-scoring AI essay and the student’s essay were critically compared and analysed. They were then tested for similarity to pre-existing literature using Turnitin, a widely used plagiarism detector. Results The student’s essay comprised 3044 words, written in 452 minutes. In contrast, the highest-marked AI-generated text comprised 3260 words, written in only 15 minutes. However, despite efficiency in time, the AI-generated essay demonstrated reduced originality, receiving a Turnitin similarity score of 46%, compared to the human-written score of 26%, indicating a greater overlap of content from existing sources. Furthermore, it was found that a single prompt was not sufficient to produce a high-quality, original, AI-generated essay; instead, a large series of instructions was required to produce a suitable review. Conclusions Generative AI can assist students by complementing their efforts, but it produces work that may lack originality. Further research is required to improve AI’s ability to generate original, in-depth content.
A 90-day district pilot
- Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
- Days 16–30 · Design. Choose one instructional unit and a specific teacher-supervised AI use. Compare it with the existing approach using a common learning objective, equivalent tasks and a shared rubric. Name an accountable owner and set the baseline, success threshold and budget before implementation.
- Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
- Days 61–90 · Decide. Measure unaided learning on a new task, factual errors, student workload and teacher review time. Disaggregate results where group sizes permit responsible interpretation. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.
The board decision brief
Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.
Attribution and reuse
Nabihah Hussaini; Susan M Astley; Adam Perrett; Elaine F Harkness. Can I Do Better Than AI: A Comparative Analysis of a Medical Student Essay Created With and Without Generative AI. Cureus (2026-02-10). https://doi.org/10.7759/cureus.103357
Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.