Even if your made-up number is accurate, how is that not infinitely preferable to a 100% probability of dieing after a natural human lifespan? I’m not being rhetorical, I’m seriously asking why a 0.18% probability of indefinite survival would not be preferable to a 0% probability of indefinite survival.
The ~78 years you live naturally is certainly less than 0.18% of the lifespan you could have if you get revived, given the assumption that it would be technologically impossible to revive you until after aging is solved.
The infinite return argument isn’t that great in general. It is especially not great when you are talking to someone who has already decided that they care about the actual probability.
The infinite return rate runs into a lot of problems:
First, someone’s utility function may be bounded. There’s some decent argument that human utility may be bounded. Note that this is not the same as having only a finite set of preferences. Thus for example, I could for any n, assuming n+1 people exist prefer “n people tortured” to “n+1 people tortured”, and still have bounded utility.
Second, even if utility can be unbounded, it isn’t clear that you can talk about infinite (rather than arbitrarily large) utilities. In particular, a lot of the standard theorems about utility curves start breaking down. To see the essential problem consider trying to distinguish “I live forever in paradise but first stub my toe” v. “I live forever in paradise.” You would need a mathematical framework that distinguishes them and yet you can still do most of your arithmetic. There are such frameworks, such as the surreal numbers, but it isn’t at all clear that you can use them to talk about utility in a useful way.
Third, my expected lifespan assuming cryonics works is not infinite. Say there’s always a finite fixed probability k greater than zero that I will die on any given day. Then the expected lifespan I will have is finite (working it out is a nice little exercise). And it isn’t like there aren’t lots of obvious things that could cut off a civilization even if it had the tech level to do nanotech revival or direct uploads. Even with those technologies, a nearby gamma ray burst could still fry everything. And aside from all the possible disaster scenarios, and wars and the like, pretty much nothing you do can deal with the (in my opinion very small) chance that we are in a simulation and the simulators decide to pull the plug.
I’ve seen reasonable estimates for expected lifespan given cryonics from a few hundred years to the tens of thousand. I don’t think we have any really good knowledge base for what the technology will be like to do anything more than an extremely rough guess at this point. But claiming it is infinite seems wrong and claiming that the expected utility is infinite seems to be more wrong.
I agree with your first two points, particularly the first, but the third seems wrong.
Third, my expected lifespan assuming cryonics works is not infinite. Say there’s always a finite fixed probability k greater than zero that I will die on any given day. Then the expected lifespan I will have is finite (working it out is a nice little exercise). And it isn’t like there aren’t lots of obvious things that could cut off a civilization even if it had the tech level to do nanotech revival or direct uploads. Even with those technologies, a nearby gamma ray burst could still fry everything.
In general any assumption of a constant per-period risk ensures doom, but we have uncertainty about whether such things exist, and can conceive of scenarios where the chance of destruction per period declines fast enough to give infinite expected lifespan. Any probability assigned to those scenarios gives infinite expected lifespan. Of course, by the same reasoning you have infinite expected lifespan even without signing up for cryonics.
Few people consider things “infinitely preferable” to other things. People may very well and very reasonably believe than keeping their money and increasing the quality of their life (or giving it to charity) on the here-and-now is better from a utilitarian perspective than a mere 0.18% chance at biological immortality.
Even immortality isn’t “infinitely preferable” to other things.
Even if your made-up number is accurate, how is that not infinitely preferable to a 100% probability of dieing after a natural human lifespan? I’m not being rhetorical, I’m seriously asking why a 0.18% probability of indefinite survival would not be preferable to a 0% probability of indefinite survival.
The ~78 years you live naturally is certainly less than 0.18% of the lifespan you could have if you get revived, given the assumption that it would be technologically impossible to revive you until after aging is solved.
The infinite return argument isn’t that great in general. It is especially not great when you are talking to someone who has already decided that they care about the actual probability.
The infinite return rate runs into a lot of problems:
First, someone’s utility function may be bounded. There’s some decent argument that human utility may be bounded. Note that this is not the same as having only a finite set of preferences. Thus for example, I could for any n, assuming n+1 people exist prefer “n people tortured” to “n+1 people tortured”, and still have bounded utility.
Second, even if utility can be unbounded, it isn’t clear that you can talk about infinite (rather than arbitrarily large) utilities. In particular, a lot of the standard theorems about utility curves start breaking down. To see the essential problem consider trying to distinguish “I live forever in paradise but first stub my toe” v. “I live forever in paradise.” You would need a mathematical framework that distinguishes them and yet you can still do most of your arithmetic. There are such frameworks, such as the surreal numbers, but it isn’t at all clear that you can use them to talk about utility in a useful way.
Third, my expected lifespan assuming cryonics works is not infinite. Say there’s always a finite fixed probability k greater than zero that I will die on any given day. Then the expected lifespan I will have is finite (working it out is a nice little exercise). And it isn’t like there aren’t lots of obvious things that could cut off a civilization even if it had the tech level to do nanotech revival or direct uploads. Even with those technologies, a nearby gamma ray burst could still fry everything. And aside from all the possible disaster scenarios, and wars and the like, pretty much nothing you do can deal with the (in my opinion very small) chance that we are in a simulation and the simulators decide to pull the plug.
I’ve seen reasonable estimates for expected lifespan given cryonics from a few hundred years to the tens of thousand. I don’t think we have any really good knowledge base for what the technology will be like to do anything more than an extremely rough guess at this point. But claiming it is infinite seems wrong and claiming that the expected utility is infinite seems to be more wrong.
I agree with your first two points, particularly the first, but the third seems wrong.
In general any assumption of a constant per-period risk ensures doom, but we have uncertainty about whether such things exist, and can conceive of scenarios where the chance of destruction per period declines fast enough to give infinite expected lifespan. Any probability assigned to those scenarios gives infinite expected lifespan. Of course, by the same reasoning you have infinite expected lifespan even without signing up for cryonics.
Few people consider things “infinitely preferable” to other things. People may very well and very reasonably believe than keeping their money and increasing the quality of their life (or giving it to charity) on the here-and-now is better from a utilitarian perspective than a mere 0.18% chance at biological immortality.
Even immortality isn’t “infinitely preferable” to other things.
See http://lesswrong.com/lw/116/the_domain_of_your_utility_function/