AI scribes for doctors are ambient tools that record the consultation, draft the clinical note, and hand it back for the clinician to review, edit, and sign. The US Department of Veterans Affairs is now expanding them after successful pilots: one vendor, Knowtex, reports more than 450,000 hours of EHR documentation saved across 79 VA medical centers and 7,000+ clinicians, with 88% adoption and 4.5 out of 5 satisfaction. For junior doctors, the after-hours charting that drives so much early-career burnout may finally shrink. The catch: you still review and sign every note, so the skill that matters most is the one no scribe can build for you.

The numbers behind the rollout

The VA data is worth reading in full, because this is the largest real-world test of ambient scribes anywhere:

  • Knowtex is used at 79 VA medical centers by more than 7,000 clinicians, saving 450,000+ hours of documentation time, with 88% adoption and an average 4.5/5 satisfaction rating across 5,100+ ratings (Nextgov, September 2026).
  • Abridge is deployed at 70+ VA sites and more than 300 US health systems, working across both the legacy and the new VA electronic health record.
  • Patients noticed. In a three-month Kansas City pilot, veteran satisfaction hit 95.8%, roughly three points higher than before scribes were introduced. Around 95% said they could trust their primary care team. Veterans must verbally consent to recording, and they can opt out at any point, including mid-visit.

The momentum is not limited to the VA. On September 30, 2026, UVA Health announced a partnership with K Health to extend AI-enabled virtual primary care and behavioral health across Virginia, explicitly aimed at cutting clinicians’ administrative time. This is no longer a pilot trend. It is infrastructure.

Why this lands first on junior doctors

Documentation burden is the single biggest driver of early-career burnout, and new doctors carry a disproportionate share of it. “Pajama time” charting, the notes finished at home after the shift, hits house officers and foundation doctors hardest.

An ambient scribe can genuinely return that time to you. My view: if your hospital adopts one, let it, but watch your habits. The first years are when you learn to turn a messy conversation into a clean, defensible clinical story. That is the skill being automated, and the one you need on the day the tool gets it wrong. My notes from the first year as a doctor keep returning to the same theme: early habits decide everything.

The part the scribe will never do for you

A scribe transcribes and structures. It does not do clinical reasoning, decide which symptom matters, or change the plan at midnight. Every output is a draft. The moment you sign it, the note is yours: the dose, the laterality, the omitted allergy. Responsibility does not transfer to the software.

The real risk is not dramatic. It is rubber-stamping: one tired day the draft looks fine and you sign without reading, and nothing goes wrong, so you do it again. Months later the habit of structured thinking that SBAR handovers are meant to build has quietly gone. The doctors who thrive with scribes will be the ones who use the saved minutes to think more, not less.

Five habits for working with an AI scribe

1. Get consent like the VA does

Explain it in one sentence and mean it: the tool records our conversation to help me write accurate notes, and you can ask me to switch it off at any time. Trust is the whole reason the Kansas City numbers moved.

2. Learn to write the note alone first

While you are still training, write notes from scratch on some shifts even with the scribe available. Convenience is not a curriculum.

3. Read every draft as if a tired colleague wrote it

Check the medication doses, the side, the plan, and what was left out. Omissions are the scribe’s weak spot: the tool writes what was said, not what should have been said.

4. Own your signature

Before you sign, ask: could I defend every line in a coroner’s court or a consultant review? If not, edit until you can.

5. Track the tool’s mistakes

Log the scribe’s clinically meaningful errors, especially early on. Complacency is the slow version of getting it wrong.

FAQ

Do AI scribes mean fewer doctors will be needed?

No. They reduce paperwork, not clinical work, so clinicians spend more time with patients. Given doctor shortages in places like the San Joaquin Valley, the need is for more doctors with better tools, not fewer doctors.

Do patients accept being recorded?

The evidence so far says yes, when consent is handled properly. In the Kansas City pilot, satisfaction rose to 95.8% with scribes in use. Veterans give verbal consent and can opt out mid-visit, which sets the standard every hospital should copy.

Will relying on a scribe hurt my training?

Only if you let it. The scribe drafts the note; it cannot teach you clinical reasoning. Juniors who keep writing notes from scratch on some shifts, and who review every draft properly, will end up better doctors, not lazier ones.

Health disclaimer: This article is for education and does not constitute medical advice. It does not diagnose, treat, or recommend any clinical tool or workflow. Clinical decisions should follow your training, local guidance, and the patients in front of you.