Fact · Dictation for work and study
In the UCLA randomized trial of two AI scribes, physicians rated clinically significant inaccuracies as occurring occasionally, for both products, on a five-point scale.
Checked against its source on October 16, 2026
In the source's own words
“clinically significant inaccuracies were noted "occasionally" on five-point Likert questions”
Ambient AI Scribes in Clinical Practice: A Randomized Trial, Lukac PJ et al., NEJM AI, 2025, Results.
Open the original (DOI 10.1056/aioa2501000)- What it measures
- The UCLA trial used a five-point Likert question for inaccuracies in each product's notes. The word occasionally is the scale label that the abstract uses.
- What it does not show
- A rating by the users of the scribes, not an audit against recordings. The abstract gives no per-product rate of errors, so it is not a count of error frequency.