The inability of wild-type HIV to spread by respiratory droplet has nothing to do with being an RNA virus. It has everything to do with whether or not the virus can avoid desiccation and stay infectious/viable when aerosolized, and whether there are receptors for HIV on the cells in the airways. Growing lentiviruses in cells that express envelope proteins with tropism for lung epithelial cells would significantly increase their ability to bind and enter these cells. If one wanted to weaponize HIV, changing its tropism is one way you could do it, God forbid.
There are no known examples of a fluid based virus becoming pulmonary. That is there is no example of a pulmonary virus that retains any semblance to a fluid based virus, indicating fluid based origins.
But even assuming that someone managed to artificially weaponize HIV into a pulmonary form, there is another significant drawback. When novel lethal pathogens are introduced into a species, there is a lethality degeneration curve where, over time, the disease becomes less lethal, then non lethal but chronic, then not even noticeable.
The lethality degeneration curve for HIV is faster and steeper than is typical. And with a significant change to its fluid based form, it is questionable whether the pulmonary form would be lethal, or even dangerous.
And this also separates H5N1 apart from the pack. Because since its discovery it has not shown an lethality degeneration curve at all, outside of a clade in Egypt, which became a little less lethal because it mutated into a pediatric form.
This was indeed scary, because children are much more prone to infection by other strains of influenza, and this would give H5N1 the opportunity to share genetic information with them. The icing on the cake is that, because of a foul-up in a vaccination program against water borne parasites, an estimated 10% of their population have hepatitis.