Summary
A little. The only randomised trial found 41 seconds a note, and only for one of the two products tested.2 Neither reduced time in the record outside scheduled hours.
Both improved how physicians felt. Take that as the actual finding rather than the consolation prize, because your burden and your minutes turn out not to be the same thing.
The work that keeps you late happens after the patient leaves. See what reaches it. Fourteen days free.
Book a demoWhat was measured
UCLA randomised 238 physicians across 14 specialties, roughly 72,000 encounters, to DAX Copilot, to Nabla, or to usual care. Nabla cut note-writing time 9.5 percent. DAX showed no statistically significant change. Neither reduced after-hours EHR time.22Lukac PJ, Turner W, Vangala S, et al. Ambient AI Scribes in Clinical Practice, A Randomized Trial. NEJM AI 2025;2(12). 238 physicians, 14 specialties, approximately 72,000 encounters. Nabla cut note-writing time 9.5 percent. DAX Copilot showed no significant change. Neither reduced after-hours EHR time. Both improved wellbeing.
No other randomised evidence exists for this category. Every other number you have seen is a vendor measuring itself, and in a survey of 121 AI scribe products, 42 percent would not tell you anything without a demo.33Kaczmarek K, et al. Trust Me, I Might be a Medical Device. Research Square preprint, March 2026. Survey of 121 AI scribe products.
The other evidence
An NHS England sponsored study found a 23.5 percent increase in direct patient interaction time and an 8.2 percent reduction in appointment length, with A&E seeing 13.4 percent more patients per shift.44An NHS England sponsored study reported a 23.5 percent increase in direct patient interaction time and an 8.2 percent reduction in appointment length with AI scribes. A&E saw a 13.4 percent increase in patients seen per shift. That is not randomised, and it measures something different. Where your time goes, rather than how much of it there is.
Both findings can be true. You can spend more of the appointment looking at the patient and still leave at seven.
Why the evening does not move
Two hours of documentation and desk work for every hour of direct patient care. Measured in 2016, and unmoved since.11Sinsky C, Colligan L, Li L, et al. Allocation of physician time in ambulatory practice. Ann Intern Med 2016;165:753-760.
Open the list at six. Letters. Results. Messages. Prior authorisations. Forms. Callbacks. The note is one item and it is the only item any ambient scribe reaches. Nine and a half percent of one item is 41 seconds. The trial result and the arithmetic agree.
What moved instead
While the category optimised the seconds, the money left through undercoding at 19 percent of E/M visits, downcoding at a reported $28,000 to $74,000 a physician, and preventable denials at 27 percent eligibility. See our essay on that.
What we changed
WA\ covers the consultation, the front desk and the patient record on one system, so the briefing opens from the chart, the letters draft from the chart and the consultation together, and the phone is answered by something that can see the room.
What we are not claiming
We have no randomised trial. Nabla does and it beat DAX. Forty-one seconds measured under randomisation is worth more than everything we can currently assert about ourselves, and we would rather say that than pretend the trial does not exist because it is inconvenient.
Availability
WA\ Clinician runs a 14-day free trial with no minimum term. Run it on a real list and look at what is still open at six. Pricing is on one page.
Frequently asked questions
How much time do AI scribes save?
About 41 seconds a note, for the one product that produced a statistically significant result. UCLA randomised 238 physicians across 14 specialties through roughly 72,000 encounters to DAX Copilot, to Nabla, or to usual care. Nabla cut note-writing time 9.5 percent. DAX showed no significant change. Neither reduced after-hours EHR time. No other randomised evidence exists for this category, and every other figure you have seen is a vendor measuring itself on its own data.
Why do AI scribes not reduce after-hours work?
Because the note is one item on a long list. Physicians spend close to two hours on documentation and desk work for every hour of direct patient care, measured in 2016 and unmoved since. Open your list at six and it is letters, results, messages, prior authorisations, forms and callbacks. The note is the only item any ambient scribe reaches, so a 9.5 percent improvement on one item is 41 seconds. The trial result and the arithmetic agree.
Does the NHS evidence show AI scribes save time?
It shows something different rather than something contradictory. An NHS England sponsored study reported a 23.5 percent increase in direct patient interaction time and an 8.2 percent reduction in appointment length, with A&E seeing 13.4 percent more patients per shift. That is not randomised and it measures where time goes rather than how much of it there is. Both findings can hold at once. You can spend more of the appointment looking at the patient and still leave at seven.
Go home when your last patient does.
Thirty minutes, your workflow, running live. Bring the work that is still open at six o'clock.
About this article. Written and published by WA\, which sells an ambient documentation product and has just told you the randomised evidence for its own category is 41 seconds. The trial, the NHS study and the Sinsky time-motion work are independent of us. The trial tested versions current in 2025 and both products have been updated. WA\ was not in that trial and has no peer-reviewed clinical trials published to date. All names are marks belonging to their owners.