He says that a) yes there are many receptors which will be triggered by the same drug at the same time, so it can have a variety of effects, b) it's possibly not worth it to take it as a preventive measure. That doesn't mean he says it shouldn't be taken to cure or at least manage debilitating disorders like depression. Cost vs benefit.

The discussion of costs/risks of treatments is usually what's missing in the casual conversations about drugs. Ketamine or psylocybin are probably the best examples in the tech community, yes they can sometimes cure a depression, or at least suppress it for a while, but they can also cause psychosis (and other non-psychiatric symptoms), so they are used only after safer options fail.

Although I'm not sure I get the concern here, sepsis is very life threatening, so paying SNRI-level risk for prevention doesn't seem like an outright bad idea. Unless the idea would be for random people to start using this drug just in case they get stabbed and go septic - then I would say this is not a good idea. But I don't think this is what the paper describes.