Received: by alpheratz.cpm.aca.mmu.ac.uk id RAA06597 (8.6.9/5.3[ref pg@gmsl.co.uk] for cpm.aca.mmu.ac.uk from fmb-majordomo@mmu.ac.uk); Mon, 18 Feb 2002 17:41:33 GMT Date: Mon, 18 Feb 2002 12:36:22 -0500 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Subject: Fwd: Building a better robot species From: "Wade T.Smith" <wade_smith@harvard.edu> To: skeptic@listproc.hcf.jhu.edu, memetics@mmu.ac.uk Content-Transfer-Encoding: quoted-printable Message-Id: <06C3D1FC-2496-11D6-88A5-003065B9A95A@harvard.edu> X-Mailer: Apple Mail (2.480) Sender: fmb-majordomo@mmu.ac.uk Precedence: bulk Reply-To: memetics@mmu.ac.uk
Building a better robot species
By Scott Kirsner, 2/18/2002
http://www.boston.com/dailyglobe2/049/business/Building_a_better_robot_speciesP.
shtml
In the spirit of reader service, some news items from the next 50 years:
White-collar workers strike to protest loss of jobs to robots; studies
find robots spend 94.6 percent less time discussing last night's ''Fear
Factor'' around the oil warmer.
Patient sues robotic surgeon for malpractice when it removes her liver
instead of spleen; bot blames buggy new Windows 2025 operating system.
Geraldo's intellect augmentation surgery derided as a stunt to boost
ratings.
After reading the new book, ''Flesh and Machines: How Robots Will Change
Us,'' by Rodney Brooks, I'm ready for the revolution that Brooks
predicts will succeed the PC and the Web: the robotics revolution.
Brooks, the director of MIT's prestigious Artificial Intelligence Lab,
says we're already seeing the early stages in the evolution of three
types of robot species.
One variety is super-sophisticated bots that operate on their own,
performing dangerous tasks like repairing highways and mundane ones like
cleaning up after us. A second kind would be remotely ''possessed'' by a
human, allowing a furnace repairman in Framingham to fix a balky valve
in a Finland basement via robot, just as today programmers in India
develop software for Silicon Valley companies. Finally, there will be
hybrid-bot humans with robotic appendages and implanted microchips.
At first, prostheses and processors will help restore mobility and treat
neurological disorders, but it won't be long before duffers are shelling
out for systems that lengthen their golf drive, and politicians are
getting chips installed that grant them perfect name-and-face recall.
Brooks' book, out this month, suggests that in the not-too-distant
future, researchers will create robots that think, feel, repair
themselves, and reproduce. Some of these next-generation machines could
rival or surpass human intellect. Brooks refers to this possibility as
the third assault on humans' special place in the universe, after those
put forth by Galileo (Earth isn't the center of the action) and Darwin
(we're not so different from animals after all).
This is a book that will spark intense reactions and generate
mind-twisting questions.
Some people won't like the notion that Brooks believes robots can have
emotions and consciousness, same as us. Others will find themselves
wondering about an inevitable robot emancipation movement: Will it be
ethical to ask robots that have personalities and even feelings to serve
as our slaves, toiling in the fields and washing our windows?
On the flip side, if they one day surpass human intelligence, will they
treat us ethically, or will we find ourselves, like astronaut Dave
Bowman in the film ''2001,'' trying desperately to pull the plug on
artificial beings that have no use for us humans?
Brooks is one of the most accomplished researchers in the field of
robotics, and also one of the few celebrities, having co-starred in the
1997 Errol Morris movie ''Fast, Cheap and Out of Control,'' which
chronicled his work building robots that mimicked insects. (The film
also garnered him fans; Brooks married MIT theater professor Janet
Sonenberg after she sent him an e-mail about the movie, though they'd
never met before.)
Brooks grew up in Adelaide, Australia, and started building computers
when he was 10 years old, including one that was unbeatable at
tic-tac-toe. His first robot, Norman, was able to wander around the
house, reacting to light and avoiding obstacles.
At MIT, Brooks began to build multi-legged robots, including one called
Genghis, which interacted with the world much as insects do, clambering
over any barriers and gravitating toward the invisible infrared halo
created by warm-blooded mammals. Genghis' actions were governed by just
a few simple rules - it didn't have any high-level operations, like
trying to map out the terrain it was traveling based on images from a
camera - but the result was a robot that seemed to exhibit life-like
behaviors and intentions.
Brooks and his students also worked with NASA to design autonomous
robots for planetary exploration, which led to the Sojourner landing on
Mars on July 4, 1997. That was the first time a robot had operated on
another planet without human control - an artificial creature relying on
its own smarts to get around.
These days, Brooks is heavily involved in running the Artificial
Intelligence Lab, spearheading its fund-raising efforts, and helping to
oversee a move to the new Frank Gehry-designed Stata Center next year.
He also serves as chairman of iRobot, a Somerville company that he
helped start that develops robots for the military, consumers, and
industrial applications.
On a recent visit to Brooks' office, he showed off a system developed by
his students as part of the lab's Project Oxygen initiative that allowed
him to speak commands and have the window blinds open or close, or turn
down the lights and start a slide presentation.
''I have a mid-life crisis every nine years,'' Brooks announced, sitting
before a coffee table in his office. ''In 1992, I shifted from insects
to humanoids, and now I've changed again, to living machines. What is
the difference between living matter and non-living matter?''
He continued: ''We're starting to build robots with things that robots
haven't had, like self-repair, self-reproduction. We want to build
robots out of sloppy material, like humans are made of. We joke that we
want to build a robot out of Jell-O. It may only last three days, but
lots of biological organisms only live that long. We'd like to be able
to build robots that get their own energy.''
That could mean robots that can find a wall outlet and plug into it when
they're running low on battery power, or it could mean robots that
actually wring energy from some sort of organic ''food,'' as humans do,
and know when they need more of it.
It can sound a bit Frankenstein-esque to someone who isn't immersed in
the field.
His response to most questions about the ethical implications of his
work is a brisk brush-off: ''You used to not be able to dissect dead
people, because of the sanctity of the human body,'' Brooks says. ''A
lot of the things we accept today as normal are things that used to make
us queasy.''
And he's steadfast that there shouldn't be limitations on robotics or
artificial intelligence research. ''It's incredibly naive to say we
should ban research in this area, because we don't know where it is
going to go,'' Brooks says.
Robots that Brooks has helped develop aren't just traveling to other
planets or skittering around like spiders. One product from iRobot
operates autonomously in oil wells; another helps the British Ministry
of Defense dispose of bombs. Earlier this month, Hasbro showed several
new robotic toys developed in conjunction with iRobot at the American
International Toy Fair in New York.
None of those bots, though, is advanced enough to get us fretting about
maintaining our dominance atop the IQ totem pole.
Brooks expects that to change quickly. He acknowledges that today's
robots pale in comparison to science fiction creations like C3PO, R2D2,
Commander Data from Star Trek, and Hal 9000. But, he writes, in just
five years, the boundary between science fiction and reality ''will be
breached in ways that are as unimaginable to most people today as daily
use of the World Wide Web was 10 years ago.''
I believe it. But robotic columnists that produce more insightful copy,
more quickly?
I've got bad news for you lazy human scribes: We're already here.
Scott Kirsner is a contributing editor at Wired and Fast Company. He can
be reached by e-mail at kirsner@att.net.
This story ran on page D1 of the Boston Globe on 2/18/2002.
© Copyright 2002 Globe Newspaper Company.
===============================This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
see: http://www.cpm.mmu.ac.uk/jom-emit
This archive was generated by hypermail 2b29 : Mon Feb 18 2002 - 17:51:18 GMT