Problem with machine learning is that it is have no grey areas. A human being doing an interview with a possible autism patient can pick up nuances, machine learning that has been trained on a select low functioning subset of autistics, possibly kids ([sarcasm]because everyone knows that autism only affect kids [/Sarcasm]) will miss the gray area and miss out on people who need the diagnosis to get help from society.
_________________
"It is far better to grasp the Universe as it really is than to persist in delusion, however satisfying and reassuring" (Carl Sagan)