Insights

Short articles on how we build clinical AI: safety, explainability, fairness and evidence.

Articles

Why our AI says 'I don't know'

A system that always has an answer is a system you can't trust. Abstaining is a feature we design for, not a failure.

How we test fairness across patient groups

Average performance can hide uneven results. We break our results down by age, sex, ethnicity and deprivation, and we'll publish what we find.