It's silly at multiple levels.
As an explanation of Fermi's Paradox it fails to explain why, if all these dead civilizations are detectable enough to get destroyed, we haven't detected any. Even if they are now extinct, their emissions must have been great enough to get them killed. So where are they?
It's very, very unlikely all of them went quiet because they learned of this out of pure theoretical reasoning. So where are their "corpses" so to speak?
And if they cannot be detected easily, because they are too far apart or emissions are near impossible to detect or recognize as evidence of intelligent life (the more likely actual explanation of Fermi's Paradox other than the simpler "they just aren't there"), then there's no risk of destruction.
Exactly. I think it's popular because it takes a difficult question and answers it with a conceptually elegant answer that has an evocative and spooky nature metaphor. Unfortunately, since it's so poorly grounded, any second order imagery built off it doesn't really add an explanatory power and usually just winds up as a tortured metaphor.
For example, Yancey Strickler's The dark forest theory of the internet blog post (which he later spun into a book) that made it so popular in think pieces like this completely misunderstands even the dark forest theory metaphor itself.
1: https://www.ystrickler.com/the-dark-forest-theory-of-the-int...