Losing 50% of species sounds bad, but much less bad than losing 50% of all mammal populations. Are we just talking about losing ten thousand obscure rodent spec

Losing 50% of species sounds bad, but much less bad than losing 50% of all mammal populations. Are we just talking about losing ten thousand obscure rodent species? I actually have no real idea what this value means, which makes me inclined to ignore it. A scientific source would be great, but the article doesn't provide any. If we're talking about "human extinction" level threats, it would probably be from massive famines, and "half of the animals are now dead" seems much more likely to cause that than "half of animal species are extinct". What is the actual percentage estimate that we would go extinct from this, and when? It matters when comparing it to nuclear war or AI. The incentive for scientists to lie seems obvious to me. They want people to support and fund their efforts (possibly for noble reasons, possibly because they just want more money). They think that simple facts aren't good enough, so they use fear tactics like "we could face our own extinction" hoping that will convince us instead. Also, the authors of these articles and the people interviewed are not necessarily the scientists either. Given that sources aren't provided, I have no idea what the scientists are actually saying.