[ExI] [ZS] [cryo] Nick Bostrom, Anders Sandberg, Stuart Armstrong to be frozen after death

[Anders Sandberg]

On 2013-06-12 07:21, Florent Berthet wrote:
> 2013/6/12 Anders Sandberg <anders at aleph.se <mailto:anders at aleph.se>>
>
>     On 2013-06-11 23:45, BillK wrote:
>
>         On Tue, Jun 11, 2013 at 11:10 PM, Keith Henson wrote:
>
>             As Anders mentioned, the payment is from life insurance
>             and as he
>             says, it's less than the cost of a dinner out a week.
>
>         Correct, but it also involves giving the life insurance payout
>         to a
>         cryonics company rather than to his estate for his family.
>
>
>     But is this a general problem? At least in my family I am the
>     least well-off: none of the others need my inheritance.
>
>
> What if the insurance payout is given to x-risk mitigation?

Depends on what xrisk mitigation we are talking about. If that money 
would reduce the probability of eventual extinction by 0.1%, sure, I 
would put the money there (I think the astronomical waste argument is 
pretty strong: the future is worth *a lot*). But realistically, it would 
be used to allow some people to think a bit more about xrisk (about a 
few postdoc man-months if we are lucky), maybe coming up with good 
ideas, which maybe get implemented and maybe have effect. Multiplying 
together all that produces a much, much smaller impact. I am not certain 
that impact is worth less than the subjective value of my life to me, 
but it seems at least possible. [*]

It would likely be more useful to spend that money on doing xrisk 
advocacy to get much more funding. Max Tegmark made the point pretty 
well in his recent talk (http://www.youtube.com/watch?v=kZDVv-MI0VU - 
this particular point starts at 41:10). The amount of funding for xrisk 
research is minuscule, and if we could get some tiny faction of the 
money wasted on truly frivolous or destructive pursuits redirected in 
this direction we would get far, far more than what we could get from 
all cryonicists switching their insurance to xrisk reduction. Now, I 
happen to spend a fair amount of my time trying to convince people about 
the importance of xrisk reduction, as well as contribute to the 
research. I have no good measure of my effectiveness, but it seems 
likely that I have a positive impact - I am already doing those 
man-months mentioned above as part of my job and overall lifestyle. It 
would hence be a really good thing if I could go on doing this for a 
long time: hence cryonics.

That argument might get the handful of xrisk researchers with cryonics 
contracts off the hook (a surprisingly large number, by the way, given 
the rarity of either pursuit) but I think there is a general efficiency 
argument too. Quibbling over individual health choices in a rather small 
population while enormous resources are going to waste is... wasteful. 
It is a bit like building managers asking people to remove AC-adapters 
from plugs in order to save energy, while maintaining floodlights to 
light up the façade at night. It is easier to point at the discrete 
individual actions than the overall system, but it is usually at the 
system level the big wins are.


[*] Fun counting exercise. Suppose we regard ozone depletion as an 
averted potential xrisk. I doubt it would have been 100% fatal: let's 
assume ozone depletion if it happens has 10% chance of wiping out 
humanity (this is on par with numbers I get for nuclear winter scenarios 
from experts). If we ignore the research necessary for its discovery 
(lots and lots of chemistry, meteorology and the work needed to launch 
the satellite that saw it), there is still a lot of research effort. 
Crutzen, Molina, and Rowland devoted at least a fraction of their 
research careers to it, and it is likely safe to assume that there was 
much relevant research outside that group. Let's assume 10 years of work 
for each of them, making 30 man-years. There are 50,000 hits in Google 
scholar for pre-1976 papers with "ozone depletion"; assuming just 1% are 
relevant and each took a month to write, that is still 500 man-months of 
work, about 41 man-years. (The 1976 Academy of Science report on 
halocarbons was written by about 15 people: if they took a month of work 
each to write their part it would already have been more than my 
cryonics insurance could pay for). So if we sum all of this up, it seems 
reasonable to say that the total effort that led to the discovery and 
recognition (not counting the later mitigation work) is on the order of 
50 man-years. So 50 man-years can buy a 10% reduction of a xrisk risk. 
If a research scientist salary is $80k, the total cost of that is $48 
million.

This is of course a debatable example, since it was in many ways an 
*easy* xrisk to fix - single cause, anthropogenic, not too powerful 
vested interests, easy substitutes - and we resolved it. The amount of 
money and effort spent on decreasing nuclear war risk is likely orders 
of magnitude larger, and has perhaps yielded less.

In any case, if my $100k were spent on ozone depletion reduction back in 
the day (I am already handwaving so much we can ignore inflation) I 
would have helped the field by 0.2%. That means an xrisk reduction of 
10% * 0.2% = 0.02%. But that assumes I picked the right xrisk to reduce: 
of all the research topics only a handful turn out to be xrisk-relevant. 
There are about 5 million researchers in the world; if each topic gets 
10-100 researchers it means there are about 50,000-500,000 research 
topics at present. We know of less than 100 xrisks, so the chance of one 
topic discovering one might be around 0.2%-0.02%. So my help, if applied 
randomly, might reduce xrisk by a factor of 4*10^-5 or 4*10^-6.

Nick's astronomical waste argument *still* multiplies this with a huge 
factor of saved future, of course.


-- 
Dr Anders Sandberg
Future of Humanity Institute
Oxford Martin School
Oxford University

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20130612/6f655d48/attachment.html>