About
A system that gets stronger every time it's caught out
The name
Why Anti Fragile
Every clinical AI system will make mistakes. The question is what happens next. In most products, a mistake is an embarrassment to be explained away. In ours, it is the raw material.
Every omission the verifier catches, and every one it misses, becomes a test case. The evaluation set grows, the bar rises, and the next version has to clear it. Errors make the system stronger instead of quietly repeating.
What we believe
Three principles
Show the gap
An answer that hides what it doesn't cover is more dangerous than one that admits it. We would rather show a clinician an uncomfortable gap than a confident answer.
Measure before marketing
We label every claim and publish our evaluations, including the ones that don't flatter us.
Clinicians decide
The verifier reviews answers. It does not make clinical decisions or overrule anyone.
Design-partner pilots
Help us build it on your cases
We are looking for a small number of hospitals, medical centers, clinicians and research groups to shape the verifier and its evaluation with us.
- ≥85%Target
of severe omissions caught
- ≤10%Target
of complete answers wrongly flagged