How confident are we that is the case?
HIV resulted from the fusion of, IIRC, three virii in a host animal.
The odds of any single such event are vanishingly small, but that stacks against how many cells per second a virus can infect, worldwide. Life is a frightening goodness-of-fit optimizer at the scale of microorganization.
In the case of HIV:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3234451/
Now, should we start taking all the viruses known to infect every species of squirrel, rat, monkey, gorilla and chimpanzee and start splicing in motifs that bind, say, the top 100 most common human cell-surface receptors in order to see what's the most dangerous virus - with the optimal mix of mortality and transmissibility (R factor) - we can create, the one best able to cause a global human pandemic? That's insane, and there's no justification for it.
[1] https://www.thelancet.com/journals/lancet/article/PIIS0140-6...