[jerf]

I mention it because it's the earliest one I can think of in our technological development. If we continue advancing, we will eventually get to that genetic diversification bit, but it seems that happens after you get the point where you can wipe out your own civilization with nukes, so that one is much less reliable.

(The Singularity paragraph is just for fun. Almost by definition, there is no reasonable way to ever collect evidence for it.)