Re: Artificial life likely in 3 to 10 years

Tim Tyler <seemysig@xxxxxxxxxxxxxx> wrote:
DK wrote:
Tim Tyler <seemysig@xxxxxxxxxxxxxx> wrote:
DK wrote:

Simply put, we are not smart enough. No artificial life any time
soon. The best we can do is to build crude natural selection emulators.

That's because the predictive power is very low. Which isn't surprising
considering the complexity of biological systems and the fact that we
can't even model a single biochemical pathway strictly. Same deal as
with climate modeling.

When Big Blue beat Gary Kasparov ten years ago, the researchers
skipped the step of modeling the human brain. You don't need
to model biological systems to be able to beat them.

You can only say that when Deep Blue writes a new and improved
version of itself. [...]

Seems to me like I can say it now, without much fear of correction:

You don't need to model biological systems to be able to beat them.

Machines can travel faster than any animal - and they use wheels,
not legs.

They can fly faster than any bird - and they use jet engines,
not muscles.

....and so on.

Except that you were taking about "beating" humans in terms of
*takeover*. All the things you mentioned above - they beat us
because we *made* them to. Without us, they are just piles of
atoms. To displace us, they either need to be smart or extremely
self-sustainable and efficient. Jet engines have neither of
these characteristics.

As for artificial life - you may have a point there. Researchers
have been pouring funds into the wrong approaches for 40 years
now - not a sign of intelligence. Creating living systems from
plausibly-prebiotic components looks pretty simple to me.

Hasn't been done and may not be ever done if it turns that
probability is very low.

IMO, it's probably happening all the time in the oceans.

However, we won't know for sure until we do it.

Yeah, it's like in the old joke - probability of it happening is 50%:
it is either happening or not. :-)

However, that enterprise is an irrelevant sideshow as far as the immediate
future of the human race goes. Companies, computers and robots
will be the non-human players there.

As of today, there is absolutely no difference between robots and
computers. That is, both are dumber than a cricket. [...]

Moravec puts current $1000 computer/brain capacity about level
with a lizard:

http://www.transhumanist.com/volume1/power.150.jpg

Bad URL. But if he really does it (which seems to be true from
googling) then he is 1) quite obviously not a biologist, 2) likely
to be insane. :-)

It is hard to compare "smartness", since they are
good at different things. The computer will trounce
the cricket at board games, but the cricket will
probably win at various "cricket" problems - like
managing long-distance jumps.

Nor they show any potential to be as efficient as microbes.

Are /we/ "as efficient as microbes"? What does it mean -
and why is it relevant?

It's not too difficult to see scenarios how microbes can beat us
- simply by being able to survive where we cannot (killing us
is an option, too). Once again: to compete with high intelligence,
you've got to be either as intelligent or extremely adapted and
efficient.

Companies ... care to outline details? I am not sure what exactly
you are thinking of. I can think of several scenarios, none of which
seems very sustainable/plausible.

From an earlier message:

``The artificial lifeform most likely to become dominant
is probably going to be a descendant of today's companies.

Companies today have a legal identity. They can
own property, and sign documents. They have their
own rights and liabilities in legal proceedings.

This immediately puts them streets ahead of the
second-most-plausible candidate, the robot.

http://alife.co.uk/essays/new_organisms/
''

Ah, this. The URL you provide does not explain how companies
will function without humans. If that does not happen, I don't see
how they qualify for artificial life any more than ant nest.

I plan to live through the takeover era.

A plan to live at least few centuries is a sign of considerable
optimism. :-)

DK


.

Relevant Pages

  • Re: Artificial life likely in 3 to 10 years
    ... to model biological systems to be able to beat them. ... Companies, computers and robots ... both are dumber than a cricket. ... Mass is an imperfect metric - since quality may not be ...
    (sci.bio.evolution)
  • Re: Artificial life likely in 3 to 10 years
    ... to model biological systems to be able to beat them. ... You can only say that when Deep Blue writes a new and improved ... Companies, computers and robots ...
    (sci.bio.evolution)