Medical schools and teaching hospitals in the United States are expanding early training to help students and incoming residents handle the daily realities of electronic health records (EHRs) and, increasingly, ambient artificial intelligence (AI) documentation tools. The push is gaining momentum as residency programs face persistent concerns about clinician documentation time, EHR-related workflow friction, and new patient-safety risks tied to AI-generated notes that require human review. Educators say the goal—starting in medical school rather than residency—is to give learners hands-on practice with chart navigation, order entry, and documentation responsibility, including how to verify ambient AI transcripts before signing clinical documentation. The effort reflects a broader shift in health information technology priorities from adoption alone to safe, competency-based use.
Context: EHR work is a core clinical task—yet training often lags behind practice
EHRs have become the operational center of modern clinical care, shaping how clinicians review history, place orders, communicate results, and document decisions. But time-motion research has repeatedly shown that EHR work and related documentation demands can consume substantial portions of clinician time and can contribute to
Frequently Asked Questions
Why are medical schools starting EHR and ambient AI training earlier, instead of waiting for residency?
Because many students and new residents encounter EHR workflow demands only after they’re already under clinical pressure. Earlier training helps learners build speed and familiarity with chart navigation, order entry, and documentation practices before residency. Educators also want to reduce documentation friction and avoid unsafe habits that can emerge when people learn the system “on the fly.”
What exactly is “ambient AI” in clinical documentation, and how does it differ from other AI uses?
Ambient AI documentation tools typically listen to clinical encounters and generate draft transcripts or note content for clinician review. Unlike decision-support that suggests diagnoses or treatments, ambient AI focuses on producing documentation artifacts. The key difference is that the output still requires human verification, because it may misinterpret intent, omit context, or include inaccuracies that could affect care if signed without review.
How are schools teaching students to verify ambient AI transcripts before signing a note?
Training emphasizes a verification workflow: cross-checking the transcript against what occurred in the encounter, confirming dates, medication details, allergies, and clinical reasoning, and ensuring orders and problem lists match the real plan. Students learn to distinguish draft language from clinically validated statements, then only sign after reviewing for completeness and correctness.
What patient-safety risks are associated with AI-generated notes, and what does “safe use” look like?
AI-generated notes can introduce errors such as incorrect clinical facts, missing key negatives, or documentation that doesn’t align with the care plan. “Safe use” means clinicians review and correct drafts, validate critical elements, and ensure documentation supports safe continuity of care. Schools aim for competency-based habits—knowing when to trust, when to verify, and when to rewrite.
Does adding EHR and ambient AI training increase student workload or distract from clinical learning?
The intent is not to add busywork, but to teach core clinical documentation tasks as part of competency development. By practicing navigation, ordering, and sign-off responsibilities early, students may spend less time struggling later. Well-designed curricula integrate EHR skills into existing clinical education so learners apply them directly to patient care scenarios.