I think the life extension stuff is probably not going to work, not because it can't (i.e. is not physically possible or so unreliable or slow not to be practic

I think the life extension stuff is probably not going to work, not because it can't (i.e. is not physically possible or so unreliable or slow not to be practical) but because all or most of the current AI researchers will die first. Maybe people will keep working on it after that but how people think can change quickly. The progress of technology (especially in the exact direction you want) is not something to be taken for granted.