I ask because I have a son with autism, and I have met leading immunologists in famous medical schools who subscribe to this hypothesis. That doesn't make the hypothesis right, but it is taken far more seriously than your comment would imply.
Second, "severe autism is really some amount of autism with very low IQ" part is also wrong in my opinion. Autism can be so severe that it makes IQ irrelevant (even in cases where it can be demonstrated to be high). The analogy here is with other mental illnesses like schizophrenia - when it is severe enough, it masks every other trait of that person, including IQ. That does not mean "severe schizophrenia is really some amount of schizophrenia combined with very low IQ". There are many severely autistic individuals who also demonstrate high IQ in specific areas (but still cannot lead anything close to a normal life), directly contradicting your statement.
I expect a little more thoughtfulness on a top comment in HN.