[ExI] Apocalypse Soon: Has Civilization Passed the Environmental Point of No Return?

[Anders Sandberg]

I was not too impressed by the paper, but it does have a point: 
theseresearchers have for a while developed a theory for detecting when 
noisy complex adaptive systems jump to other equilibria, and they are 
making the case that this might be applicable to the global ecosystem.

The problem is that their argument in Nature is mainly based on one set 
of models where they see things happen when a kind of percolation limit 
is passed: this can be turned into a prediction for when bad things will 
happen, but the uncertainties involved are big enough to give plenty of 
wiggle room - or nasty surprises.

Their *good* argument, about slowing of fluctuations near bifurcations, 
is not really used. This actually gives a way of checking whether we are 
entering the danger zone, and is where I would have been gathering data 
series to bolster the first argument if I had been writing the paper.

One interesting issue that might be worth following up is to check their 
percolation argument and see if it applied to economies as they 
integrate or globalize. Ought to be a good way of checking it.

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