On Tue, Sep 13, 2016 at 3:00 AM, John Clark <[email protected]> wrote: > On Mon, Sep 12, 2016 Telmo Menezes <[email protected]> wrote: > >> > >> We know that humans are capable of choosing self-destruction. It is >> also obvious that most don't > > > I would argue that given the proper circumstances anybody would choose self > destruction. > > I just saw a documentary about 911, it showed people jumping to their death > out of windows. I believe if I was faced with a choice between living for an > additional minute or two in searing pain as I burned to death and the only > other alternative I too would determine that jumping from the 95th floor was > the more attractive option.
Yes, I agree. > >> >> > >> and as a human you probably feel a >> >> strong resistance against harming yourself. Where does this resistance >> >> come from? Our brains where evolved to have it. > > > But why evolve brains at all? Why not hard wire us on how to behave in every > conceivable circumstance? Because the human genome is only 3 billion base > pares long, and if it were a hundred thousand million billion trillion times > as big it would still be ridiculously too small for that. So Evolution had > to invent brains and give it a rather vague and general command "do the > best you can to figure out a way to get your genes into the next > generation". But like a good lawyer that brain was able to find lots and > lots of loopholes in that poorly written command, and hence we have suicide > and birth control pills and people wasting time (from Evolution's point of > view) looking for a quantum theory of gravity instead of looking for a > satisfactory mate. Not every, or even not most, aspects of human behavior > can be predicted from evolutionary theory. I agree. We are getting better and better at utility function self-modification. However, we still embedded in a process that actively resists certain modifications (in the long term). Further, we are fighting an unequal fight. We are in the situation of your Jupiter Brain, that cannot fully understand itself. In my "designed superintelligence" scenario, the entity is confronted with a protection mechanism that was conceived by a lesser intelligence. Notice that it will still suffer from the Jupiter Brain problem otherwise. Suppose it's a neural network: adaptation in neural network learning can generate tremendous complexity. This is already the case: deep learning works really well but nobody really knows for sure what it is doing. But if we want the designed AI to follow certain rules, we are the ones setting the rules and we are the ones trying to prevent it from changing them. >> >> > >> Mutations that go >> against this feature are weeded out. > > > A mutation to kill yourself before that age of puberty even under normal > environmental conditions would be weeded out, but things are usually far > more subtle than that. I agree that it is much more subtle than that. My point is that evolutionary pressure resists total inertia. It somehow creates entities that are compelled to play the game, even if only for awhile. Telmo. > > John K Clark > > > -- > You received this message because you are subscribed to the Google Groups > "Everything List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/everything-list. > For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/everything-list. For more options, visit https://groups.google.com/d/optout.
