Re: the robot scenario

John Blanco-Losada (jbl@clark.net)
Sat, 16 Nov 96 10:14:52 -0500


Max M wrote:

>SCENARIO
>1) Computers wil obtain intelligence/autonomity either by traditional top
>down AI methods (like the CYC project), or because all of the autonomous
>processes on the net wil somehow reach a state of conciousness through
>shear complexity. Maybe they wil achieve autonomity by some other mean it
>doesn't really matter. What matters is that they will able to support and
>develop themself without human assistance.
>...

Coincidentally enough, not 15 minutes after I read this message I ran
across the following Reuters story posted yesterday:

> READING, England (Reuter) - A robot used the internet to
>program a robot in the United States Friday without human
>intervention, in what British scientists said was a world first.
> ``This is a major breakthrough for machinekind,'' said Dr
>Kevin Warwick, Professor of Cybernetics at Reading University,
>west of London.
> During the experiment, one robot at the university's
>cybernetics laboratory taught another at the State University of
>New York (SUNY) how to move around its environment.
> ``This is the first time a robot has ever programd another
>robot,'' Warwick told reporters.
> The experiment aimed to show how behavior learnt by one
>robot could be passed on to another thousands of miles away via
>a transatlantic internet link.
> ``We're trying to show that computers can learn the same way
>as humans, by communicating learning,'' said Dr David Keating,
>who started the research program six years ago and devised the
>idea of the internet link-up.
> The tricycle-like robots, which are six inches high, and
>weigh 21.2 ounces require no remote control or human input once
>they are switched on.
> The robot in Britain taught the robot at SUNY how to
>determine its position relative to other objects using its
>ultrasound sensors, like bats. The robot was then able to learn
>by trial and error using its own microprocessors.
> The robots are even capable of homing in on and docking with
>a recharger when its batteries are low, the scientists said.
> Asked whether he was concerned if computers would be able to
>overtake mankind, Warwick said, ``We're all right for the next
>10 years but after that it starts to get worrying.''
> ``It could have apocalyptic consequences for mankind after
>the next 15-20 years,'' he added.
> The two robots used in the experiment both belong to Reading
>University and form part of a troupe called the ``seven
>dwarfs.''

-----------------------------------------------------------------------
John Blanco-Losada "You must be the change
jbl@clark.net you wish to see
http://www.clark.net/pub/jbl/jbl.html in the world." - M. Gandhi

Do you believe in Macintosh?
Join the EvangeList at <http://www.evangelist.macaddict.com/>
-----------------------------------------------------------------------