Some studies find that "schizophrenia runs in lower average IQ families, autism runs in higher average IQ families", from which some suggest that the schizophrenia-vs-autism distinction is grounded in the IQ distinction – see e.g. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4927579/ – but I think that kind of research is very speculative and needs to be taken with a large grain of salt

Other researchers suggest the boundary between autism and schizophrenia is weaker than is commonly supposed. Common diagnostic tests for ASD perform poorly in distinguishing it from schizophrenia; there is heightened prevalence of each condition among those diagnosed with the other; autism started out as a new name for childhood-onset schizophrenia (in fact the word "autistic" was originally coined to describe schizophrenia), and there is still a lot of dispute about what is the difference between autism and childhood-onset schizophrenia (the former is much more common than the later–subclinical hallucinations are common in children, but clinically significant psychotic symptoms are quite rare)

Nobody really knows, not even the experts.