"the person would then be dead" seems to pretty clearly imply that there was a person involved. In the case where a corpse poofs into existence from nowhere, there doesn't seem to have ever been a person involved. I conclude that "Because the person would then be dead" doesn't apply to the case where a corpse poofs into existence from nowhere. So I'm not sure why we would need to further specify anything here.

All of that said, the whole approach of counting deaths as negative utility seems to me to be rescuing the wrong part of the original nonconsequentialist claim in the first place.

It's clear that one consequence of increasing the human population from 1 billion people to 7 billion people is that many more people die per unit time, but it doesn't follow from that fact that we should reject increasing human population on consequentialist grounds. (It might be true that we should so reject it, but even if true it doesn't follow from that fact.)

It seems that the part we would want to rescue from a consequentialist POV is the idea that more life-years is good, so any act that reduces expected net lifeyears is bad... and also, perhaps, the idea that more life-years/person is good, so any act that reduces expected net lifeyears/person is bad.

This would also render all concerns about how we define "death" moot.