I'm not so sure.
For one, most of us know the dangers of extrapolating past the end of data points, so we shouldn't base longevity changes on an extrapolated graph.
Secondly, there are signs that longevity has peaked. There was an article a couple of weeks ago that got bandied about in the Australian press (which I can't find now, darn it) that life expectancy is now a little shorter than it was a decade ago. From memory, it was USA data, so perhaps a US reader will remember it.