Agentic education · Learning science

Actionable companion · Assessment & reasoning

From AI evidence to responsible practice: Artificial intelligence in undergraduate medical education clinical skills curricula: a scoping review of implementations since 2022

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not findings or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paperSearch paper agentCC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Higher education. Health-professions education findings do not establish effectiveness in K–12 classrooms.

Original authors’ abstract, reproduced under the linked source license:

Purpose To systematically identify and synthesize peer-reviewed literature describing implemented AI innovations within undergraduate medical education clinical skills curricula from January 2022 through January 2026. Method The authors conducted a scoping review querying PubMed and Scopus, supplemented by SciSpace as an AI-assisted citation discovery tool. Eligible studies described utilizing AI to deliver the clinical skills curriculum in innovative ways (e.g., instruction in history-taking, communication, clinical reasoning, clinical documentation, OSCE/simulation assessment). We extracted data into standardized templates and thematically sorted to characterize how AI-assisted instruction was being implemented across educational objectives. Results From 1,130 initial records, 39 studies met inclusion criteria. AI-assisted instruction clustered into eight thematic categories: LLM-Based Virtual Patient and Clinical Simulation Systems (n = 19), AI-Augmented OSCE and Simulation Assessment Tools (n = 6), Embodied and Robotic AI Clinical Simulations (n = 4), AI-Supported Procedural and Technical Skills Training (n = 3), AI-Assisted Clinical Documentation and EHR-Based Skills Training (n = 2), Multimodal Analytics for Skills Assessment (n = 2), Educator-Facing AI Case Authoring and Simulation Design Tools (n = 2), and AI-Supported Clinical Reasoning and Tutoring Tools (n = 1). Publication activity concentrated heavily in 2024–2025, with virtual patient applications representing the dominant category. Conclusions AI implementation in clinical skills education has accelerated substantially since 2022, with large language model-powered virtual patient simulations emerging as the predominant application. Current implementations primarily position AI as a supplementary formative tool rather than a replacement for established pedagogical approaches. Robust evidence regarding long-term educational impact remains limited, indicating need for rigorous longitudinal evaluation alongside continued innovation.

What kind of evidence is this?

Supporting AI learning science · Evidence synthesis

Stage: Higher education. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Study methods · source passage

We conducted a scoping review to characterize how AI-enabled tools have been implemented in undergraduate medical education clinical skills curricula and skills-adjacent training domains from January 1, 2022, through January 2026. This review followed the PRISMA extension for Scoping Reviews (PRISMA-ScR) guidelines. The review aimed to identify AI use as an adjunct to clinical skills training (e.g., history-taking, physical examination, communication, clinical reasoning, documentation, OSCE/simulation training and assessment) and to categorize where implementation is most concentrated across the current literature.

Limitations and uncertainty · source passage

This scoping review has several limitations. Despite a comprehensive search strategy, relevant AI-assisted implementations may have been missed, particularly non-English publications, institution-specific initiatives, or emerging conference outputs. In addition, the literature search was completed in January 2026, and the rapid pace of innovation in this field means that new studies may have appeared since that date.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

  • Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.
  • Co-design accommodations with learners and specialists. Check access and error burden across groups without inferring subgroup effectiveness from small or unreported samples.

A 90-day district pilot

  1. 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.
  2. Days 16–30 · Design. Choose one disciplinary task. Require a first attempt, a documented AI consultation, a critique of the output and an independent explanation. Align the comparison group and scoring rubric with the same objective. Name an accountable owner and set the baseline, success threshold and budget before implementation.
  3. 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.
  4. Days 61–90 · Decide. Blind-score reasoning and accuracy on a new task without AI. Distinguish model examination scores, student perceptions and actual learner gains; only the last directly tests educational benefit. 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

Birpartap S. Thind; Daryoush Javidi; Lisa M. Schwartz. Artificial intelligence in undergraduate medical education clinical skills curricula: a scoping review of implementations since 2022. Frontiers in Digital Health (2026-06-10). https://doi.org/10.3389/fdgth.2026.1830254

Companion prepared 2026-11-01. 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.