Hacker News Reader

They're not even real cases, so there's nothing to verify.

This study is so far below the standard of evidence for medicine it's comical.

It's just another of those "Look how this general intelligence can distinguish between the data on the training set!" that get spammed everywhere nowadays.

Multiple doctors, with consensus. Can still be wrong for very complicated cases, but 99.99% of the time its correct.

Well you can invert the approach. So you:

1) Take a given ailment

2) Take all the symptoms that it produces

3) Choose some at random

4) Construct a synthetic patient case with those symptoms where you know exactly what the problem is

They did the equivalent of unit tests but for doctors.... seems sensible