July 27, 2009 1:17 PM

How Close Is "Terminator"-Like World?

By
CBSNews
(CBS)  "Terminator Salvation," the latest in the series of movies about titanic clashes between man and machine, was unleashed on big screens in theaters this weekend.

And, reported CBS News correspondent Cynthia Bowers on The Early Show Monday, the future depicted in the films may be closer to fact than fiction.

The "Terminator" movies tell a horrifying tale of what might happen if machines made by men turn on them, a science fiction spectacle that may not be all that far-fetched, according to the Brookings Institution's P.W. Singer, author of ""Wired for War: The Robotics Revolution and Conflict in the 21st Century."

"We may not be at the 'Terminator' world right now," he says, "but we already have 12,000 ground robots today in Iraq and Afghanistan." Not to mention as many as 7,000 pilot-less predator drones.

Technology, says Bowers, has come so far, so fast, the U.S. Navy even commissioned a study on the ethics of military robots - a study that resulted in a ""black box"-type warning, saying in part, "Autonomous military robots that will fight future wars must be programmed to live by a strict warrior code or the world risks untold atrocities at their steely hands."

While many scientists are quick to point out that robots usually do jobs too dull, dangerous, or dirty for humans, and insist there's no evil intent, other researchers are concerned about what could happen when, as Bowers puts it, "scientists start programming machines with the most human trait of all, a desire for self-preservation."

Robots are already having artificial intelligence programmed in which, says Bowers, "begs the question: When a machine starts thinking on its own, does it ever stop?"

Copyright 2009 CBS. All rights reserved.
Add a Comment See all 29 Comments
by Mixolydian8X11 May 29, 2009 4:38 PM EDT
"'When a machine starts thinking on its own, does it ever stop?'"

. . . Just turn on Rush Lim-bot, I'm sure you could get half of 'em to stop thinking on their own.
Reply to this comment
by Mixolydian8X11 May 29, 2009 4:36 PM EDT
"robots usually do jobs too dull, dangerous, or dirty for humans"

. . . Hmm, last time I checked, the undocumented immigrant population isn't robots.
Reply to this comment
by sean7phil May 28, 2009 12:47 PM EDT
Some people just like being skeptical.

A robot doesn't have to be 'conscious' or highly advanced to become dangerous-- there already very dangerous robots being used in the field by the military.

But its a complex issue-- because more combat robots mean less of our soldiers will be killed.

Although potentially more of the enemy in and also civilians in Third World countries could be killed in battles fought using American automated or robotized force.
Reply to this comment
by rational_1 May 28, 2009 12:06 PM EDT
A true artificial intelligence is still a far-fetched dream; all we have now is more and more sophisticated versions of that calculator you have sitting on your desk.
Posted by willcad at 8:51 AM : May 27, 2009

Interesting post - and I'd have to agree with you. For example, there isn't a computer out there with the cognitive skills of an insect. Ever seen the behaviours exhibited by wasps and bees? Sure many are instinctive (hard wired), but I don't think there is a computer out there that could deal with its environment nearly as well as a bug does on a routine basis.
Reply to this comment
by willcad May 27, 2009 11:51 AM EDT
I disagree that we are centuries away (possibly decades). Look at how far we have come with computers in the last 20 - 30 years. I can't remember the exact formula, but according to Moores Law, micro processors (which are the brains of a computer) double in capacity every few years. This has been pretty constant the past 30 years or so. They can already program computers to "think". Look at the computers that play chess against grand champions. I think we are 20 years or less away from creating the first CPUs that can totally think for themselves. Where we go from there, who knows?
Posted by tramplers at 2:28 PM : May 26, 2009
-------------------------------------------------------------------------------------------------------

Moore's Law actually states that the number of transistors per square inch of integrated circuit would double every two years. He wasn't talking about "capacity" (whatever you meant by that), or about processor speed or amount of RAM or amount of storage - he was talking about the physical limits of the manufacturing process used to create integrated circuits. As technology progresses, new ways to etch small transistors onto silicone chips are found, increasing the number that can fit in any given space. This has the effect of increasing the processing power of any given chip.

Computers that play chess aren't really "thinking"; they are simply processing huge numbers of if/then/else statements. Ask any of those computers to tell you who will win the next American Idol, and they won't be able to respond - because they are not really thinking, they are computing, which is an entirely different process.

A true artificial intelligence is still a far-fetched dream; all we have now is more and more sophisticated versions of that calculator you have sitting on your desk.
Reply to this comment
by zippiez May 27, 2009 10:31 AM EDT
This just a bedtime story based on impossible assumptions similar to a story about Congress writing laws benefiting the country as a whole.

See! Its an interesting thought experiment, but the assumptions forming the basis are impossible.
Reply to this comment
by tramplers May 26, 2009 5:28 PM EDT
Machines that can actually think, feel, make judgments, or have any sense of self-awareness are decades or centuries in the future. And when they do appear, the greatest threat they present will be that they WON'T kill humans when they are ordered to do so.

I disagree that we are centuries away (possibly decades). Look at how far we have come with computers in the last 20 - 30 years. I can't remember the exact formula, but according to Moores Law, micro processors (which are the brains of a computer) double in capacity every few years. This has been pretty constant the past 30 years or so. They can already program computers to "think". Look at the computers that play chess against grand champions. I think we are 20 years or less away from creating the first CPUs that can totally think for themselves. Where we go from there, who knows?
Reply to this comment
by adt13t May 26, 2009 3:23 PM EDT
spank a nerd just for fun
no tinker toy boy
no adult diapers in the stratosphere
no astro nuts
no nasa research money wasted on monkey kidneys and lungs
no tinker toy boy
bad nerd no battery.
robotic commands and communication tech 101
what we need are actual adults at the helm in this finite biosphere
utilizing proper expenditures
How about a super hubble ish competition
lose the propulsion fantasy at this time
tinker toy boy
spank a nerd
or maybe an astrophysics major master/no/*****/boy/bation
string theory stupidiety
charlatan masquerade
sell out ego
anti tist
junk science
adult diapers in space what a stinkin grotesque waste
nasa funding operative 101.
spank a nerd just for fun
starting with astro physics
broken string out of tune theory
Reply to this comment
by displeased May 26, 2009 11:09 AM EDT
How Close Is "Terminator"-Like World? How Slow A Newsday Is It When This Passes For A News Story?
Posted by taxchurches

Slow news days can be a good thing.
Reply to this comment
by casionova May 26, 2009 10:44 AM EDT
As an autonomous English nerd robot, I must point out the misuse of the term, "begs the question", which means, "assumes the answer within the question". The reporter should have said "leads to the question" or "prompts the question". The T-800s are on their way to her office now.
Posted by ghostfighter-2009

Oh no! The grammar terminators are coming! ;o)
Reply to this comment
See all 29 Comments
.
Scroll Left
Scroll Right More »
CBS News on Facebook