I really don't want to spoil your dinner but the chances of millions of dead per year are fairly high, and that's up to 2100, not many models look beyond that (the uncertainty is too large). The extremes of those models (things that could happen but with a much smaller likelihood) are enough to make you miserable.
The good news is that there are similar possible outcomes that things won't be that dire.
250K dead / year by 2050, 5 million / year or so by 2100.
The tough part is that such models have a lot of uncertainty and that it is very well possible to be alive but to be utterly miserable. I'm normally fairly optimistic about our ability to influence our lives but on this one I'm not, this will happen and we're along for the ride, it's a rearguard action at this point where we can try hard to minimize the damage but not avoid it completely.
When we think how many of those deaths - and the suffering leading up to them - could be changed by reducing obesity, reducing sugar intake, increasing exercise, reducing air pollution, reducing smoking, increasing vitamin D levels, reducing alcohol consumption, reducing stress by any means (city noise to hypercompetitive workplaces to ragebait media), improving road safety, and "we" don't act on most of that. And then how much more we would have to act to hold or reverse climate change to save comparatively fewer people, it seems ... not a bad idea, but a sadly unlikely one.
(Incidentally, 67 million -> 72 million is about a 7% increase in deaths. Which is in the region of the excess deaths in 2022 which correlate with COVID vaccination in 2021 in European countries. A study from a Norwegian University finds a correlation of 1% increase in population vaccinated in 2021 -> 0.1% increase in excess deaths in 2022. Not caused by COVID which they excluded, not caused by a bounceback after a year of fewer deaths which they accounted for, and as far as the authors can tell not caused by delayed access to health care which was widespread and not linked to vaccination levels in different countries; and if that's roughly in the region of the expected climate change deaths, you can see that not only are we not doing much about this, we are barely even talking about it in public, which doesn't bode well for the climate change equivalent: Dr on YouTube citing studies and evidence and not conspiracy theories: https://www.youtube.com/watch?v=iyo2UNQcdpQ )
Yes, obviously the number of expected deaths is ~ the number of people alive / average lifespan.
But the number of people dying due to climate change is a difficult to pin down number. Do you or do not include for instance people dying in wars when the proximate cause for the war was climate change or mass migration? Some believe you should, others say that wars have always been fought. So it's a tricky problem for many intrinsic reasons and it is an even trickier problem because there are so many ways to interpret the data.
All this adds up to us doing nothing until it is way too late.
> 250K dead / year by 2050, 5 million / year or so by 2100.
Can that be fully offset by increase standards of living and improving healthcare/medicine? 250k is not a lot statistically on a global scale (and I really doubt the capability of these models to accurately predict what's going to happen in 50+ years)
> The tough part is that such models have a lot of uncertainty and that it is very well possible to be alive but to be utterly miserable.
I'm not saying it's going to be tough for a lot of people in some places. But I don't really see how for humanity as a whole it can be much more than a slowdown than potentially can be fully compensated by further technological progress. Of course increasing economic inequality can likely only make it worse rather than better
Mass die offs have happened historically but nothing short of the black plague is in line with the range of possibilities before us.