> Why is someone on HIV antivirals if no test ever confirmed them to have HIV?

Mu. If it was confirmed, but not "confirmed by laboratory criteria according to country definitions and requirements", then they do not meet the diagnostic criteria (interpreted literally). Suppose, for instance, that there was a procedural error that might have messed up the diagnosis (so is forbidden by regulation), but in this case didn't mess up the diagnosis.

I can produce as many of these literally-correct, deliberate misinterpretations as you like. They have no bearing on actual medical practice.

> which definition is correct?

Which definition of "carbon atom" is correct? Our definitions have, for 200 years, been sufficient to distinguish "carbon atom" from "not carbon atom", but those definitions have changed significantly in that time. Autism is that category into which autistic people fall, and into which allistic people do not fall, which is distinguished from several other categories with which it is often confused. (The ICD-11 spends way more words on distinguishing autism from OCD, Tourette's, schizophrenia, etc than on defining it directly.)