Diagnosis (general) · AI doctors and chatbots
AMIE: Google's diagnostic chatbot research
Google's AMIE is a research AI that takes a medical history by text chat. In a study published in Nature in 2025, it was rated as more accurate than primary-care doctors on 159 simulated cases with actors as patients. That study used no real patients; a later real-patient study is tracked as TR-016.
How far away is this?
Simulated patients only in this paper. A later, closely supervised feasibility study with real patients (The Lancet, October 2026) is tracked separately as TR-016.
Where is it approved?
| Region | Regulator | Status | Notes |
|---|---|---|---|
| United States | FDA | Not approved | Research system, not offered to patients. No openFDA 510(k) record found (checked 16 Oct 2026).[2] |
| European Union | CE mark / EMA | Not approved | Research system, not offered to patients (Google calls it a research chatbot). We found no CE mark; EUDAMED not searched.[2] |
| United Kingdom | MHRA | Not approved | Research system, not offered to patients (Google calls it a research chatbot). We found no MHRA registration; PARD not searched.[2] |
| Australia | TGA | Not approved | Research system, not offered to patients (Google calls it a research chatbot). We found no ARTG entry; the ARTG search was unavailable.[2] |
A research system, not offered to patients anywhere. 'Not approved' is based on that and on finding no regulator records, not on a regulator statement. "Unknown" means we haven't verified it, not that it's unavailable. Approval in one country isn't approval in another.
What is it?
AMIE is a large language model trained to ask questions like a doctor and suggest possible diagnoses.
What's the evidence?
In a randomised, double-blind text-chat study with 159 case scenarios and 20 primary-care physicians, AMIE had greater diagnostic accuracy and was rated better on most measures by specialists and patient-actors. Google built AMIE and ran the evaluation.[1]
Evidence stage: Early research · #c1
The authors say the results should be read with caution: doctors were working in an unfamiliar text-chat format, and patients were simulated.[1]
Evidence stage: Early research · #c2
What does it mean for me?
Be careful treating any chatbot as a doctor. General chatbots aren't AMIE, and AMIE itself has only been tested with real patients in one published study we know of, a small and closely supervised one. If a chatbot tells you something is minor but you feel very unwell, seek care anyway.
What could I ask my doctor?
- I looked up my symptoms with an AI tool. Can we go through what it said?
- What symptoms mean I should come back sooner?
These are general starting points, not a recommendation to choose any particular test or treatment.
Change history
- : First published.
- : Updated after independent fact-check and clinical review
- : Linked to the newer real-patient study (TR-016); regulator notes reworded to match.
- : Review label corrected: this page was not clinically reviewed. Earlier labels saying 'independent fact-check' and 'clinical review' were wrong; checks so far were AI-assisted source checks only.
General information only, not medical advice. Talk to your doctor about your situation. AI-assisted source check. Not clinically reviewed.
References
- Tu et al., Nature (2025). nature.com (2025). Peer-reviewed study. https://www.nature.com/articles/s41586-025-08866-7
- FDA record or statement: United States. blog.google. Regulatory or manufacturer record. https://blog.google/innovation-and-ai/technology/health/amie-clinical-study-lancet/
Publisher shown as the website hosting each source. Dates are given where the source label includes them.