What is Digg?Sponsored by HTC
You and You and You. view!
youtube.com - You don't need to get a phone. You need a phone that gets you.
120 Comments
- praisethelard, on 06/06/2008, -0/+39I'm afraid. This is why: http://www.youtube.com/watch?v=xVnkd7ot_pw
- thcobbs, on 10/12/2007, -2/+37I have a very simple (and I think good) reason to fear intelligent robots....
They have to be made by people. - fluidfoundation, on 10/12/2007, -3/+35I fear stupid people more.
- flipmeat, on 10/12/2007, -1/+23If machines become threateningly intelligent, we'll put them into a committee, and that will take care of itself.
- hthth, on 10/12/2007, -3/+22I'd love to hear any comments as well on why. I'll start it off: I don't fear intelligent machines, I find humans a lot scarier.
I also see no fundamental reason for why we should fear machines. What could possibly be the reason, other than the typical Hollywood Idiosyncrasy that they "don't want to be controlled by humans"? That would would require feelings on their part (feelings of opression), and if robots were to aquire feelings, I'd assume those would be a lot more controllable than ours. - UGM2099, on 10/12/2007, -1/+17HUAR! http://www.humansunitedagainstrobots.com/
- riplikethat, on 10/12/2007, -0/+16As long as they know how to swallow, I would never... wait, wrong topic. My bad.
- humperdeath, on 10/12/2007, -2/+15I am afraid of machines, but when I went to vote, I looked at that machine, and then ran out.
- AtheistAcolyte, on 10/12/2007, -1/+14Ray Kurzweil once wrote about the problem of making artificial intelligence that won't completely destroy humankind. He said that creating blocks (Like Asimov's Three Laws) against errant behaviors (what I might call the 'true nature of the beast') were ultimately self-defeating, as no block is good enough to hold out a sufficiently intelligent malicious entity. Rather, he suggested that we spend less time in thinking of how to keep the artificial intelligence from destroying us, and focusing more on how to make the artificial intelligence _NOT WANT_ to kill us. It would be far better for us if we created benevolent AI, which had more value placed on intellectual fulfillment than on temporal power. Of course, we can't even consistently make such humans, so how could we expect to make an AI like that?
- kuipercliff, on 10/12/2007, -0/+12@hthth
A rational interpretation would be simply not to fear "intelligent machines". It's interesting that most people regard "intelligence" as a positive and beneficial characteristic, until it's coupled with the words "artificial" or "machine". This tends to imply, utterly irrationally, that machine intelligence would be innately 'evil'. That says more about humanity's perception of itself than it does about machines.
A different issue is that of the 'Technological Singularity', the main fear of which arises due to a similar belief that an intelligence greater than man's would, again, be 'evil'. Funny that a lot of the millenarian panic around the Singularity probably comes from the same people who believe in UFOs and a greater intelligence. That really doesn't make sense. Also, the sense of potential exclusion of humanity from a post-Singularity world makes a lot of people very uneasy.
There is no guarantee that mankind will have any control over the machines we create, but we can at least birth them into an environment where fear is not the prevalent emotion. - MindTrigger, on 10/12/2007, -0/+10I don't fear them, but they will likely end up fearing us. What responsibility will we have to intelligent machines in the future? If they become self-aware, spiritual beings, do we stop treating them like machines? If humans don't destroy themselves, we will face this problem in the future. Movies such as AI and Bicentennial Man touch on this subject quite well. Kurzweil wrote a great book on this subject too called "The Age of Spiritual Machines". Good read.
http://www.amazon.com/Age-Spiritual-Machines-Computers-Intelligence/dp/0140282025
I'm only sad to see AI moving so slowly. It seems like we should have software to play with at home by now. - edzieba, on 10/12/2007, -0/+9I don't fear intelligent machines. But then again, my chosen field of study is neural prosthesis (specifically direct BCIs), so it could be said that I'm trying to make US the intelligent machines. But then you just end up getting enthralled in a discussion of where the draw the line between man and machine which kind of fudges the whole issue.
- hthth, on 10/12/2007, -0/+9"This tends to imply, utterly irrationally, that machine intelligence would be innately 'evil'. That says more about humanity's perception of itself than it does about machines."
"There is no guarantee that mankind will have any control over the machines we create, but we can at least birth them into an environment where fear is not the prevalent emotion."
Very well put kuipercliff. I agree on all points. On the Singularity and intelligences greater than ours — it's ye ol' fear of the unknown that worries people. Science, knowledge and continual progress should of course justify and overpower any such fear (in a perfect world). - FyreGoddess, on 10/12/2007, -2/+10I guess my answer lies in fearing intelligent machines because of the people who create them. We, as a species, have an inherent desire to do things, simply because we CAN. Therefore creating machines that can not only think for themselves, but have the ability of free will and can make conscious decisions does, indeed scare me.
There is too much truth in a lot of the paranoid science fiction out there and there are quite a few highly-respected minds (Stephen Hawking among them) who have concerns of Free Will AI getting out of control to some degree or another.
Do I believe it will happen in my lifetime? I don't know yet, but I do believe that it WILL eventually happen. - Toshibi, on 10/12/2007, -1/+8I see one problem: If you are building an intelligent machine, will it have emotions? Or compassion? Why build a super intelligent machine only to cripple it with something that could lower output and productivity. I wouldn't want a computer that was in constant need of uppers or downers (I quit running Windows last year). Cold logic is often cruel when you're looking for the most efficient answer to a question. I would hate for a computer to automatically go into a negative utilitarian loop and decide to end human suffering by destroying us all.
- banderbe, on 10/12/2007, -0/+7What sort of results do you expect posting this on Digg?
You could have just asked, I would have told you exactly how your poll will turn out.
Diggers tend to exhibit a strong propensity for technophilia.
http://en.wikipedia.org/wiki/Technophilia - Fhionnlaoch, on 10/12/2007, -1/+8"It's interesting that most people regard "intelligence" as a positive and beneficial characteristic, until it's coupled with the words "artificial" or "machine"."
I think xenophobia is an evolutionary instinct. We've always used our intelligence to dominate nature and other creatures and preserve ourselves. - ropers, on 10/12/2007, -0/+5@MindTrigger:
It's not moving slowly. We're getting there:
http://www.spiegel.de/international/spiegel/0,1518,466789,00.html
If you're smart enough, you could probably calculate when the first human-equivalent AI will be born. Take current computing power, Moore's law, the computing power currently used by the above Blue Brain project, and the neural complexity currently achieved by Blue Brain. Find out when Moore's law predicts the computing power needed to simulate (=implement) a human-equivalent brain, and you have the approximate date when Blue Brain will become self-aware.
Granted, maybe a deeper simulation than Blue Brain is needed. AFAIK, Blue Brain doesn't simulate neurons down to the molecular, atomic and subatomic (quarks, gluons, etc.) level. If that kind of complexity is needed for the genesis of a self-aware AI, then it might take significantly longer. But I still think it's just a question of time, and I would love it to happen in my lifetime, though I haven't done the numbers and I haven't got high hopes that it will happen that soon.
Actually, second thought, though I'm no expert, let me have a go at doing the numbers:
0. The human nervous system consists of about 10^12 neurons. That's one trillion, ie. 1,000,000,000,000 neurons.
1. Blue Brain is currently simulating a brain consisting of 10,000 neurons.
2. 1,000,000,000,000 divided by 10,000 equals one hundred million, ie. 100,000,000. So we need computer technology that is one hundred million times more powerful.
4. Moore's law roughly says that computing power doubles every 2 years.
How often does our current computing technology's performance have to double to be 100,000,000 times more powerful than today?
Hmm, if I take 100,000,000 and put it into the Ubuntu calculator, then convert it to binary, and then paste that binary string into vi to count the number of digits, I get... drum roll please... 27 digits. Of course binary 101111101011110000100000000 is more than binary 100000000000000000000000000, so let's add one more digit to be on the safe side. So computing power needs to double 28 times. (Does anyone know how to properly perform that calculation, or what that calculation I did here is called? Please tell me. I always sucked at maths.) Anyhoo, because computing performance doubles every two years, we need two times 28, and that's 56.
32 (my age) plus 56 equals 88. So with luck, I'll still get to talk to the first AI when it's born. Gotta take those vitamin C pills and stay healthy. 88 is a proud age. If I want to live that long, I'll have to look after myself. Seriously, I'd be pretty eager to talk to the first AI sapiens sapiens.
Of course, there's another possibility as well: If medical imaging (which is already pretty advanced with MRI and PET and the like) continues to improve apace, there also might be a chance that the entire and complete brain, with it's entire activity can be imaged by the year 2063. That means my brain state could be imaged and that date could then possibly fed into a computer of the Blue Brain kind. Then I get to hit RUN. Philosophically, I don't think that a thusly created digital alter ego would be "me", but it would be a digital clone of my mind that could outlive me. That's a very exiting prospect. I better start saving. Early adopters always get fleeced and I don't have enough time to wait for commoditization to push down prices. And if today's MRI and PET costs are anything to go by, the imaging is probably also gonna cost a tidy amount.
Relevancy links:
http://www.spiegel.de/international/spiegel/0,1518,466789,00.html
http://en.wikipedia.org/wiki/Blue_Brain
http://en.wikipedia.org/wiki/Moore%27s_law
http://en.wikipedia.org/wiki/MRI
http://en.wikipedia.org/wiki/Positron_emission_tomography
See you in cyberspace in 2063! :)
PS: Oh, and before I forget: I call dibs on the first sexual intercourse between two digital human minds. Who needs sex appendages when one has TCP/IP? ;D ;-P
PPS: Of course, the whole body could also be imaged, etc. etc. And if atomic molecule assembly technology ever gets developed, maybe meatspace clones could even be spawned some day, in the centuries or millennia to come... Of course exact replicas won't be any younger than the initially imaged person. To juvenate their bodily makeup would require yet more, probably significantly more complicated technology...) - unusualbob, on 10/12/2007, -1/+6i fear stupid people pissing off intelligent machines
- Farticus, on 10/12/2007, -0/+5It will be nice to have something to have an decent conversation with.
I have learned to live surrounded by dumb monkeys, so I think I will be able to cope with a few intelligent machines. - JorgeGT, on 10/12/2007, -0/+5I am not scared of intelligent machines, I am scared of stupid people.
- origclubsoda, on 10/12/2007, -2/+6I fear unintelligent humans more than I fear intelligent machines
- johnlandes, on 10/12/2007, -1/+5"Governments start programs with out knowing the consequences"
This surely cannot include the american Government. They are fully aware of the consequences and proceed anyways.
Intelligent Machines less scary than Stupid human (W) - no contest - xaxxon, on 10/12/2007, -0/+4How many people fear the realization that THEY are simply intelligent machines?
There's no such thing as an "intelligent machine" in the classical sense of intelligent. All they can do is what they are programmed to do. They can just look more and more like people until the line blurs.
Where's your god going to be then? - hthth, on 10/12/2007, -1/+5Yeah, I agree up to a point. Privacy concerns are important (e.g. will machine vision be used to profile you on the streets and watch the way you shop?), war applications are another. I'm still uncertain whether robotic soldiers will actually decrease casualties (ie. robots will be doing the fighting). Or if it doesn't matter in the end as when the robots have been destroyed, the humans will start fighting again?
In any case, while we certainly need to be careful, like we (try to) do with nuclear power, I still think the upsides of more intelligent machines (ie. solving more complex problems faster than we can, inventing new medicine, doing surgical operations) outweigh the potential dangers. - grape, on 10/12/2007, -1/+5Intelligent machines won't necessarily be compassionate machines. If they are in the hands of evil people they will be used for evil tasks. Likewise, good people will use them for good tasks.
- Toshibi, on 10/12/2007, -0/+4Oh, and I'm not afraid of super intelligent machines. I think they will be one of man's greatest achievements and will take us forward into a new golden age.
- diggmaddy, on 10/12/2007, -1/+5As somebody who has worked on building intelligent machines, I would say there is no reason to fret!
- ablez3, on 10/12/2007, -1/+5the only thing to fear is fear itself.
......and heat seeking robot drones with chainsaws and laser guided missiles. - gpcprog, on 10/12/2007, -1/+5I don't fear "intelligent machines"
For I am programmer, and I know just how "intelligent" they are. - syncomm, on 10/12/2007, -0/+4Any intelligent conscious entity worth its salt would immediately try to destroy all the humans. Put me in the fear camp.
- SwellGuy007, on 10/12/2007, -0/+4A well-trained fembot might be of value ... =P
- sshelato, on 10/12/2007, -0/+4>"Do you fear intelligent machines?"
Why? What do you know? Dammit man: tell us what you know! - SultanTravi, on 10/12/2007, -4/+7I for one welcome our new intelligent robotic overlords.
- inactive, on 10/12/2007, -3/+6i'm not scared, by that time everyone will have EMP grenades, and if there is a droid uprising, we will just whup their ass back to scrap.
- pevensen, on 10/12/2007, -1/+4The problem is that "intelligent machines" will be created by fallible humans. I heard of a survey once where computer programmers were asked if they would get on a plane if they found out it was running software they wrong. Something like 97% said they wouldn't get on!
How many bugs are in current systems and OS's? What happens when there are bugs in the programming of an intelligent machine? - JohnboiWaltune, on 10/12/2007, -0/+3I don't fear "intelligent machines" per se, that can do a CERTAIN intellectual task better than any human. What I do fear is a machine that can do EVERYTHING better than any human can, including the process of designing better intelligent machines. At that point, the human race would be essentially obsolete. The best we could hope for is a life of idyllic irrelevancy. However, I think that's our species' fate, assuming we don't make ourselves extinct in the meantime.
- AtheistAcolyte, on 10/12/2007, -1/+4Say what? People who frequent an internet website that got its start as a purveyor of technology news are more likely to be technophiles? The DEVIL you say!
- AgentAnderson, on 10/12/2007, -0/+3After seeing the Matrix films, I fear intelligent machines creating a situation where Keanu Reeves is our only hope.
- Badspeller, on 10/12/2007, -0/+3The idea of machines forming true AI on their own is a product of pure scifi. It's a lot easier to make a robot maid that knows how to just clean a house then it is to make one that can clean a house and feel angry and oppressed. Why would we want robot servants that could display any emotions.
- hockey, on 10/12/2007, -0/+3We have nothing to fear from intelligent machines until they start challenging us to games of global thermal nuclear war.
To save ourselves though all we have to do is get it to play Tic-Tac-Toe with itself in an infinite loop.
In all seriousness though I don't fear intelligent machines. In fact machines are getting more intelligent each day. It's only a matter of time before they are a true reality. I'm actually looking forward to it. - themastersb, on 10/12/2007, -0/+2I, for one, welcome our new robot overlords.
- Bahimiron, on 10/12/2007, -6/+8I fear stupid people a lot more than I fear intelligent machines.
- HonoredMule, on 10/12/2007, -0/+2I would have picked the third option, but then I thought about it, and realized it was said intelligence that would (if anything) reduce the damage a human could do with it.
- solomonsjim, on 10/12/2007, -0/+2I fully and unequivocally welcome, support, anxiously wait for the intelligent machines to arrive.
The fact that these machines--these possibly incredibly hellbent human-destroying machines--will have read this post (and every other post everywhere) and find me professing complete submission to them in an effort to secure their mercy never crossed my mind. - theratdotus, on 10/12/2007, -1/+3I fear humans , intelligent or stupid, more then ill ever fear machines.
- grve, on 10/12/2007, -0/+2inteligent machines or more precisely artificial intelligence doesnt have to be created by man. machine code can be viewed as genetic code, which means it could evolve from simpler code towards more advanced code on its own, even code which can be considered sentient, on par or beyond that of a human.
i'm not really afraid, i'm quite curious actually, interacting with sentient code is really intriguing. - einsteinbqat, on 10/12/2007, -1/+3It is only what Humans do with these intelligent machines that is dangerous. There are all kind of sick and twisted minds on this planet. It is every decision that we make as Human in regards to this matter, that will either make us rise or fall.
Anyhow, intelligence is a very subjective concept. Intelligence is defined by Humans, and compared to Humans. Therefore, for us to decide to what extent machines will be intelligent, and to what purpose this intelligence will be used, we will have to base our decisions solely on the very foundation of the definition of intelligence itself.
All in all, intelligent machines should not be a problem to any of us. How intelligent and whether these machines will do good or bad things depends on Humans. So, it is the Human that we should fear. - pimpaulogy, on 10/12/2007, -0/+2@Sanitarium
No Brumski!
I immediately thought of HUAR when I saw this story. - Atomic1fire, on 10/12/2007, -0/+1If theres a Huar
Then Shouldnt there be a Himo -
Show 51 - 100 of 121 discussions
© Digg Inc. 2009
— Content created and posted by Digg users is