Google AMIE study in The Lancet: AI chatbot aided clinician prep
A study led by Google and Beth Israel Deaconess Medical Center (BIDMC), published in The Lancet's main journal, tested Google's research-grade diagnostic AI chatbot AMIE with 98 patients before their primary-care visits, with clinicians monitoring every interaction in real time. No conversation needed to be interrupted under preset safety criteria; clinicians said AI-generated summaries helped them prepare in 75% of scenarios, AI output changed their diagnostic thinking in more than half of cases, and AMIE's differential diagnoses matched physicians' final diagnoses 90% of the time. The authors say larger trials are still needed to evaluate patient-facing AI at scale.
It is an early real-world clinical test of a diagnostic AI chatbot's safety and assistive value, offering preliminary evidence for regulators and for designing larger patient-facing trials.