[ExI] Planetary defense

[Anders Sandberg]

I am working on my speech for the IAA Planetary Defense conference 
("From Threat to Action"). It is such a cool name; the fact that it is 
held in the Bucharest Palace of the Parliament (just google this 
monstrosity) fuels my imagination of broadshouldered science-generals 
standing around a giant holodisplay screen moving missiles and lasers 
into the right orbits to meet the invaders... Of course, the real topics 
will be far more mundane - photometry, orbit estimation, mission 
planning, international coordination and so on. But it is real work on 
reducing a global catastrophic risk, which is terribly cool.

I would like to hear your input: what are the best approaches for 
dealing with the "billion body problem", the fact that  human 
rationality when it comes to risk tends to be fairly... problematic. How 
do we handle the bias that as long as no disaster has happened yet 
people underestimate the risk, that planetary defense is the ultimate 
public good, and that people tend to treat big threats as belonging in a 
fiction category that are fun to think about but not act on?

And from a transhumanist perspective, assuming accelerating 
technological progress, might it not be smart to wait when you detect a 
threat in X years time, since in a few more years we will have far more 
tech to deal with it? How do you make a rational estimation of when you 
should strike, given uncertainties in how anti-threat technology will 
develop plus the (known) increase in difficulty in deflecting threats later?

Another interesting aspect is that power-law distributed disasters with 
probabilixy P(x) = x^-a (x is disatster size, a>1 is an exponent) have 
infinite expectation is a<2 - sooner or later one is going to be much 
larger than you can handle. Even for a>2 much of the expected losses 
over time come from the extreme tail, so it is rational to spend more on 
fixing the extreme tail than stopping the small everyday disasters. But 
this will of course not sit well with taxpayers or victims. Should we 
aim for setting up systems that really prevent end-of-the-world 
scenarios even at great cost, or is it better to have mid-range systems 
that show action occasionally? Even when they fail utilitarian 
cost-benefit analysis and the MaxiPOK principle?

And watch the skies: *
http://www.universetoday.com/85360/take-a-look-huge-asteroid-to-fly-by-earth-in-november/
*

-- 
Anders Sandberg,
Future of Humanity Institute
Philosophy Faculty of Oxford University