Viruses evolve to spread as efficiently as possible, what happens to the host after they have spread is from a viral evolution point of view utterly irrelevant. So they can and do evolve to be more lethal. Maybe it takes a bit longer before you die but that's perfectly fine.
How is the norm for viruses to continue to become more lethal? That does not align with anything I've ever heard. That link doesn't even claim that, and that's a baseless "fact check" done by a mathematician acting as a false authority.
The norm is for viruses to end up somewhere in the middle, but chances are that the lethality is still very high. The whole idea that viruses evolve to be less lethal simply isn't true and very much outdated.
> That does not align with anything I've ever heard.
That is possible, it is a pretty persistent thing but it has been dealt with decisively since the 80's:
https://www.researchgate.net/publication/224043344_Coevoluti...
In a nutshell: it is likely that viruses (and other parasites) change once they have made their jump to a new species. This change is mostly random, selection pressure will lead to maximizing for R0, which may or may not lead to a lower level of lethality but this is by no means guaranteed. Some outliers may evolve to become more lethal, some other outliers may evolve to be (much) less lethal, but for the bulk an intermediary level of lethality (which can still be very high) is the normal outcome.
This is experimentally confirmed over and over again over the timescales that we have been doing these.
The shortest way in which you can condense all this is: "to a virus it is replication that matters, not what happens to the host after it has replicated". Of course viruses don't have any conscious plan but the evolutionary mechanism selects for (more) successful reproduction so it will seem like that from the outside.