Independent, peer-reviewed evaluations of ambient AI documentation, alongside the frameworks we use to hold clinical AI to a defined standard. We publish what we learn, including where the findings are complicated for us.
All articles
MBRU, Dubai
A qualitative study of clinician and simulated-patient perspectives, with four implementation lessons and the consent questions unique to mental health.
Gold Coast HHS
Clinician acceptance and patient acceptability of an AI scribe across allied health disciplines, in a mixed-methods hospital study.
The DHF 2026 keynote on what purpose-built clinical AI must entail as capability grows: a defined, countable standard for safety.
Lyrebird's measurement approach: binary, countable error classifications across five dimensions, with human judgement integrated throughout.
Purpose-built clinical AI
Lyrebird is the clinical AI platform for Australian clinicians. We back independent evaluation and publish what it finds, because the standard for clinical AI has to rise with its capability.
