Re: An essay I just wrote on the Singularity.

From: Perry E. Metzger (perry@piermont.com)
Date: Thu Jan 01 2004 - 16:47:27 MST


Tommy McCabe <rocketjet314@yahoo.com> writes:

> --- "Perry E. Metzger" <perry@piermont.com> wrote:
>>
>> Tommy McCabe <rocketjet314@yahoo.com> writes:
>> > True, but no disproof exists.
>>
>> Operating on the assumption that that something
>> which may or may not
>> be possible will happen seems imprudent.
>
> It seems like a very reasonable idea that what can be
> done, by dumb evolution, in a few gigabytes of DNA can
> be done in programming code by humans.

So one can have AI. I don't dispute that. What I'm talking about is
"Friendly" AI.

> And if you have Friendly human-equivalent AI,

You've taken a leap. Step back. Just because we know we can build AI
doesn't mean we know we can build "Friendly" AI.

>> > If anyone thinks they
>> > have one, I would be very interested. And there's
>> > currently no good reason I can see why Friendly AI
>> > shouldn't be possible.
>>
>> I can -- or at least, why it wouldn't be stable.
>
> Then please, by all means, show me the proof.
>
>> There are several
>> problems here, including the fact that there is no
>> absolute morality (and
>> thus no way to universally determine "the good"),
>
> This is the postition of subjective morality, which is
> far from proven. It's not a 'fact', it is a
> possibility.

It is unproven, for the exact same reason that the non-existence of
God is unproven and indeed unprovable -- I can come up with all sorts
of non-falsifiable scenarios in which a God could exist. However, an
absolute morality requires a bunch of assumptions -- again,
non-falsifiable assumptions. As a good Popperian, depending on things
that are non-falsifiable rings alarm bells in my head.

>> that it is not
>> obvious that one could construct something far more
>> intelligent than
>> yourself
>
> Perhaps we truly can't construct something vastly more

Stop. You cut my sentence. I don't doubt that dumb forces can build
intelligences -- we're an example of that after all. I said:

>> that it is not obvious that one could construct something far more
>> intelligent than yourself and still manage to constrain its
>> behavior effectively,

and don't edit my words that way again if you want me to reply.

>> and still manage to constrain its behavior
>> effectively,
>
> You can't 'constrain' a transhuman.

And so, I don't believe we can guarantee that you can create a
Friendly AI in the process of creating a superhuman intelligence.

>> that it is not clear that a construct like this would be able to
>> battle it out effectively against other constructs from societies
>> that do not construct Friendly AIs (or indeed that the winner in
>> the universe won't be the societies that produce the meanest,
>> baddest-assed intelligences rather than the friendliest -- see
>> evolution on earth), etc.
>
> Battle it out? The 'winner'? The 'winner' in this case
> is the AI who makes it to superintelligence first.

How do we know that this hasn't already happened elsewhere in the
universe? We don't. We just assume (probably correctly) that it hasn't
happened on our planet -- but there are all sorts of other planets out
there. The Universe is Big. You don't want to build something that
will have trouble with Outside Context Problems (as Ian Banks dubbed
them).

>> Anyway, I find it interesting to speculate on possible constructs
>> like The Friendly AI, but not safe to assume that they're going to
>> be in one's future.
>
> Of course you can't assume that there will be a
> Singularity caused by a Friendly AI, but I'm pretty
> darn sure I want it to happen!

I want roses to grow unbidden from the wood of my writing desk.

Don't speak of desire. Speak of realistic possibility.

>> The prudent transhumanist considers
>> survival in wide
>> variety of scenarios.
>
> Survival? If the first transhuman is Friendly,
> survival is a given,

No, it is not, because it isn't even clear that there will be any way
to define "Friendly" well enough. See "no absolute morality", above.

-- 
Perry E. Metzger		perry@piermont.com


This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:43 MDT