[ExI] Brain emulation, regions and AGI [WAS Re: Kelly's future]
rpwl at lightlink.com
Sat Jun 4 16:31:49 UTC 2011
> On Sat, Jun 4, 2011 at 4:29 PM, Kelly Anderson wrote:
>> I think this is all more a matter of training or upbringing if you
>> will, rather than leaving some magical module out. There are people
>> who are wonderfully non-violent, but it's not because they don't have
>> some physical module, but because they were taught that violence is
>> bad, and it sunk in.
>> Admittedly Richard did not say if this module was a software or
>> hardware module. And we probably differ a bit on how much is
>> "programming" vs. "training".
> There are some threats that cannot be negotiated away.
> The AGI must have the resources to get violent when the circumstances demand it.
> Could be tricky deciding when and what level of violence is needed.
Ah, that is a different thing, I believe, than an aggression module.
So, consider a person who has never in their life experienced feelings
of rage or anger (trying to approximate, here, an AGI with no aggression
module). Now, in a situation in which interstellar invaders turn up and
make it clear they want to set up an Intergalactic Walmart on Earth, and
they will need to erase all native life first, that non-aggressive
person might still decide that it will be necessary to nuke the invasion
force on the way in.
That would not be an aggression-driven action. No need for the agg.
module. And there is no reason to believe that a person who felt
overwhelming aggression toward the invaders would be better at building
the defence force (quite the opposite, perhaps).
I cannot think of any circumstances where violent (or potentially
dangerous) motivation modules would be *required* for the AGI to
function properly. People often quote the invading space aliens
scenario I just described, but other than that I know of none.
More information about the extropy-chat