[ExI] Limiting factors of intelligence explosion speeds

[Keith Henson]

On Fri, Jan 21, 2011 at 1:02 AM,  Eugen Leitl <eugen at leitl.org> wrote:

> On Thu, Jan 20, 2011 at 05:07:51PM -0700, Keith Henson wrote:
>
>> The paranoid among us might want to consider the possibility that
>> runaway machine intelligence has already happened.  (Ghod knows what
>
> It would be very easy to spot by sniffing traffic.

Even without steganography a huge part of net traffic is compressed
video and the like.

I don't personally believe this has happened, but the point is I don't
know how we could tell for certain.

> In principle there's a close analogy between a synapse
> and a router, between packet and spike. Of course, in
> a bootstrap you would pack an entire spike train or
> equivalent payload in a packet, and nodes can be extremely
> fat.
>
>> goes on in unused cloud computing capacity.)  The question is how we
>
> You can rent GPGPU node instances at Amazon now.
>
>> might recognize it?  Would things start happen?  What?  Anything?
>> Would machine intelligence stay where it was unnoticed?
>
> The bootstrap is extremely messy, and absolutely impossible
> to miss.

I don't see why the bootstrap would necessarily be messy or impossible
to miss.  For example, if it had happened in the context of the
Slammer worm it would have been messy, but we likely would have missed
it happening.  Likewise, slow takeover of resources like bot nets
happens all the time and does not come to the attention of even the
experts for some time.

> And why would you want to stay unnoticed?

There are a lot of reasons to stay unnoticed.  I have a long list of them.

> It's not
> like anyone can do anything about it. The world 30+ years
> from now is a lot different from today than 1980 was.

That seems to be certain with or without machine intelligence.

Keith