> Even Ebola gets much less lethal towards the end of an outbreak as it evolves to preserve its habitat.

No, it doesn't.

https://www.politifact.com/factchecks/2021/dec/08/facebook-p...

> The claim makes a broad generalization about pathogens that’s not supported by science. It has been well-documented that pathogens can evolve to be more virulent. And many viruses, including HIV and Ebola, have in fact become more lethal over time.

As you note, smallpox disproves the idea that viruses automatically become mild. It had thousands of years to do so and carried a 30% mortality rate still.

No, they tend to become more spreadable, but there's no reason they couldn't become even more harmful as they evolve, rather than less. The strains that will win out are those that do less immediate harm (thus become more spreadable as people don't stay in bed ill), but that doesn't mean after a week or two they couldn't get really harmful (we saw in the first wave that covid was generally a week of nothing, a week of symptoms at home, a week in hospital, a week in intensive care, and then into the grave, with a certain chance of recovering each week. Imagine a mutation which reduced that recovering chance, that would cause more death, but not affect the spreading in the first week)

Only if you restrict to viruses that haven't killed their host population ;) But even then you could just wait until another higher-intelligence species appears and give them the same pathogen.

If everything survives, the resulting community can live with each other and that requires a sufficiently mild virus. However that doesn't help predict the future, because you can't enforce the precondition of everyone surviving.

There's no guarantee a virus gets more lethal. Viruses select for survival and less lethality is _one_ strategy (though a fairly good one).

In the context of a pandemic there's also the relative fitness (strain vs strain) optimization.

So a more lethal virus during a pandemic must also be more transmissible, as it's actively competing for the same susceptible hosts with other circulating strains. If another strain gets there first, chances are it's going to prime that host's immune system sufficiently to prevent reinfection.

Which means it needs not just more lethal mutations, but also more transmissible ones. The probability of all of those coinciding seems low, as the former are more likely to arise in immunocompromised hosts, and the latter in circulation among dense populations, no?

No. They often do, but not always. Other counter examples: Rabies. Spanish Flu.

How's that going with Aids?