What is Digg?Sponsored by Dragon Age: Origins
Follow the Dragon Age: Origins development team on Twitter view!
twitter.com/DragonAge - EA presents BioWare's new dark fantasy epic Dragon Age: Origins. '9/10' from Game Informer.
85 Comments
- jefallbright, on 10/11/2007, -4/+18Artificial Intelligence, robotics, life extension, ethics, "the singularity", and the future of humanity. Whether you agree with their positions or not, these issues are rising on a wave of technology, and it would best to see what's coming BEFORE it hits you.
- inactive, on 10/11/2007, -2/+14Damn! There goes my plan to design an angry robot who feed on human brains.
- Simcom, on 10/11/2007, -4/+16I for one welcome our new A.I. overlords.
- MagicBobert, on 10/11/2007, -1/+13Holy *****, the guy at 1:25 has got a sweet beard.
- philrenaud, on 10/11/2007, -0/+11woahhh
dugg for that guy's beard - crowsattack, on 10/11/2007, -1/+10Can we have open source video drivers first please!
- bysin, on 10/11/2007, -5/+13Someone watches too many movies.
- jwl23, on 10/11/2007, -4/+11Wow! I see you've really put a lot of thought and study into the issue. Thanks for the contribution!
- liveforever, on 10/11/2007, -5/+11Friendly AI would definitely be better than AI that didn't care about us.
Dugg to protect the future of the human species... - rlpowell, on 10/11/2007, -5/+11I've been a regular donor to the SIAI for a couple of years now; after much research I consider them our best route to the Singularity and (more importantly) a nice world to live in *afterword*.
-Robin - hello2usir, on 10/11/2007, -2/+7EMP grenades > robots
- Foma, on 10/11/2007, -1/+6Well, you just described everything from apocalyptic evangelical Christianity rackets to scientific research into climate change, biodiversity, seismology, and anything else that involves possibilities of really ***** things happening. Everyone needs money to keep doing whatever they do, so the reason you cite for this organization seeming like a cult is not a very good one.
If you evaluate the actual evidence they have for prioritizing their research in the way they do, and you find it lacking, then that's a separate issue. The fact is that sometimes, really ***** things *do* happen; so if you find that people are providing good evidence to be cautious in a certain area, then it's counterproductive to dismiss them just because most others who talk about such world-changing possibilities happen to be crackpots. - allenu, on 10/11/2007, -3/+8WTF? This is creepy. I just finished watching this video mere seconds ago, then loaded up Digg/Videos and saw it listed. Is the Singularity here already? :-D
Also worth watching is Ray Kurzweil's talk at the Singularity summit: http://youtube.com/watch?v=9PWXrnsSrf0 - healeychris1, on 10/11/2007, -6/+11The AI research community today tends to resemble a bunch of skilled cabinet makers running around trying to build the first house. Building a house, however, involves understanding many concerns that simply have no good expression in the language of cabinet-making. Having considered this, SIAI exists to ensure that we begin to explore these previously unrecognized and unaddressed AI-related concerns in a deliberate fashion, before we might suffer the risks inherent in blundering ahead with a limited understanding of potential impacts.
Like the bystander uneasily shifting to-and-fro in the shadow of the cabinet-makers' wooden tower, I don't expect our society's AI researchers to know enough to build a perfect house, but I do expect them to have a basic framework that will make sure their house doesn't end up on top of me! SIAI is the only organization I'm aware of that is seeking such a framework with which to safely model our pursuits of AI, and they need our awareness and support! - inactive, on 10/11/2007, -2/+6PEBKAC!
- struldbrugg, on 10/11/2007, -0/+4The "you only use 10% of your brain" thing is an urban legend: http://www.snopes.com/science/stats/10percnt.htm
- rjgutzeit, on 10/11/2007, -2/+5Only with artificial intelligence are so many people so confident in the insolubility of a problem they know nothing about. Why do people who have never opened a textbook on AI feel comfortable dismissing its goals as impossible? The fact that all attempts so far have failed isn't sufficient explanation: Have you ever heard anyone with only a passing interest in math who never got far past high school calculus declare that the Riemann Hypothesis will probably never be proven, and if it is proven we can be sure it won't happen in the next 100 years, and that anyone who suggests it might be solved in five is talking pure nonsense?
Perhaps because with the Riemann Hypothesis, nothing bad happens if you're wrong. This isn't always the case. Imagine how reasonable it must have sounded when public opinion said, "We will probably never extract power from nuclear reactions. In the unlikely event we do, it won't be for a very long time, maybe 100 years. When it happens, it will be very hard and expensive and ineffective at first, and probably won't be useful for anything until after many decades of research to bring it in competition with other forms of producing energy. The idea that we could extract enough energy out of this to destroy an entire city in less than ten years is utterly absurd." Whether or not AI is the next big dangerous thing nobody saw coming, this same mentality will be there to ensure the world gets destroyed by gray goo or airborne Ebola or whatever we invent next. - inactive, on 10/11/2007, -1/+3Isaac Asimov's Three Laws of Robotics:
1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey orders given to it by human beings except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.
Later, he also added the "zero-eth law": A robot may not harm humanity, or, by inaction, allow humanity to come to harm. - rjgutzeit, on 10/11/2007, -1/+3My sentiments exactly. I'm a bit amazed anyone thought encouraging positive comments on digg would produce anything other than, "Hi, I'm an SIAI donor, and I give money for reasons foo, bar, and qux. You should give them money too," thus making the organization look more zany and cultish than it already does.
Makes me regret a little that I'm sending them money. I just really hope SIAI can eventually save its image from its own fans. - Simcom, on 10/11/2007, -4/+6yea, first one didn't go over so well. :)
- realcrunchyfrog, on 10/11/2007, -1/+3The haters commenting here need to go take their ritalin, if you can sit still long enough to view the video and think of what they're trying to do... you might find it not only laudable but quite brilliant. This is important work and it needs to be done.
An anlysis of the natural laws that accidentally gave rise to the intelligence that allows you to read this is in order. And once we suss those mechanisms out we would to well to refine them, speed them up and let them work themselves into something far greater than we could ever hope to become.
I know this scares you, but they're talking about doing it in a controlled manner. I think it's awesome and I pledge my support... - E3L1, on 10/11/2007, -1/+2pfft, they're not even giving SHODAN a chance...
- acorwin42, on 10/11/2007, -3/+4Better safe than sorry. Or at least, better that someone is paying attention to the possible implications of AI.
- struldbrugg, on 10/11/2007, -1/+2http://gdayworld.thepodcastnetwork.com/2007/05/17/gday-world-238-eliezer-yudkowsky/
- rhinopig, on 10/11/2007, -1/+2http://www.youtube.com/watch?v=I7_Old1Ntuc here is a counter-argument of sorts by a shirtless guy. (not argument against the singularity, just an alternate path I guess)
Haha, yeah that guys beard is awesome. - inactive, on 10/11/2007, -1/+2very true; Asimov himself gives many examples in his own books about possible failures of the "Three Laws". But no discussion on the subject is complete without them.
- neurobionetics, on 10/11/2007, -0/+0My apologies, I just noticed the profile doesn't seem to have a space for an E-mail.
If you are interested in joining the TSN core activist team, please contact me at: TSNtransnational@transhumanism.org
Thank you! - neurobionetics, on 10/11/2007, -0/+0It is wonderful to see the mission of SIAI shared in this popular medium.
If anyone in school or under 36 reading this far into the comments postings is interested in joining a "core activist team" to promote this and related efforts to fellow students and youth, please contact me through the E-mail on my profile.
The Transhumanist Student Network (TSN) is poised to become a more professional outreach program with sponsorship and guidance from the transhumanist community. We need students and youth who are are excited about this opportunity to help promote SIAI and other transhumanist projects to their peers to join the core team and expand our base of support!
http://www.transhumanism.org/campus
- Ben - rjgutzeit, on 10/11/2007, -1/+1What movies have ever depicted AI as genuinely benevolent on that scale? The only one I can think of where an AI did anything as helpful as curing cancer and ending world hunger is Colossus: The Forbin Project, and in that the benevolence was just a ploy to earn trust, take over the world, and go back to being evil like movie AIs always are.
Occasionally you see nice-but-impotent AIs, like Data in Star Trek, who are never smarter than humans in any non-superficial way. Data's decisions on important issues aren't any more often right than any other human member of the crew. Data can calculate the roots of large numbers or reprogram a computer with superhuman speed, but his supposedly superhuman intelligence falls short of allowing him to do anything that would benefit human society on the same scale that human intelligence benefited the apes from which we descended.
Movies have to stray from reality whenever reality doesn't make for an interesting story. Star Trek just wouldn't be Star Trek if Data was simply always right about everything, instantly saw solutions to problems that would have otherwise puzzled human geniuses for generations, was able to persuade anybody into doing anything, and routinely employed methods as incomprehensible to us as things like machine guns, lawsuits, and political speeches are to dogs. - jenniwink, on 10/11/2007, -1/+1Wow. A lot of big names have joined the SingInst crew. Kudos. BTW, peeps, Yudkowsky's papers are some of the best reads on the net.
- josephjah, on 10/11/2007, -0/+0There is a lot of merit to what they have to say... This stuff is quickly approaching us, and it is a much better idea to at least to understand what s going on rather than just ignoring it or laughing at it with little or no research under your belt.
This issue is far too important to just push under the rug people! There are individuals actively trying to develop AGI systems right now. It is a fact that computers are gaining proficiency in areas that were once deemed exclusively human.
It is kind of disgusting that people can make such ridiculous comments about such an important issue, to me it shows a complete lack of understanding in the area. I would advise anyone who plans to make a comment in this area to do at least a little research into AI before you make your comment, because no one enjoys uninformed statements... but that is far too much to ask here. - Naikokujin, on 10/11/2007, -6/+6Some of the greatest minds of our time working for our immediate future.
Worth every bit of support we will give them. - scotherns, on 10/11/2007, -1/+1The best chance we have of surviving the increasingly dangerous problems in the world is to apply more intelligence. Frankly, humans do not seem suitably equipped to do this alone. We need to build something that has the potential to become smarter than us, but if we attempt this, then it is crucial that what we build is Friendly. This is the most important project in the history of our species, and deserves our utmost attention and support.
- datagod, on 10/11/2007, -3/+3So I guess somebody is just standing there with their finger on some big "off" button, just in case the AI decided to nuke us?
- kevinosborne, on 10/11/2007, -1/+1heh meanwhile http://digg.com/offbeat_news/84_of_women_in_survey_say_they_ve_met_new_men_by_ass_grinding_in_club has hit 500 diggs in under three hours
sweet, sweet agony - soulforged, on 10/11/2007, -1/+1Unfortunately the three (or four) laws of robotics aren't adequate to ensure a safe Singularity. You can find some good arguments here: http://www.asimovlaws.com/
Therefore, we really need brilliant people to focus on these issues and as far as I'm aware, SIAI is the only organization that does this. - kevinosborne, on 10/11/2007, -2/+2as an oldtime /. & digg veteran as well as a transhumanist/singulatarian, reading these comments has been absolute gold.
siai proponents: waffle, hand-wave, 'the oracle has spoken' etc
digg freaks: grr spam, 'eat my mansquirt', wtf, etc
it's like seeing my wife and mistress meet eachother randomly only to find they are oddly repulsed by eachother
did i mention this is gold? :-) - youngdigg, on 10/11/2007, -1/+0I had a really good look at the video. Pretty certain that all the people in it are aliens. Especially the guy 20sec into the video. His head/neck ratio is all wrong.
- Pragmatica, on 10/11/2007, -1/+0We need more forward-looking organizations such as this. Strong AI will likely arrive during all our lifetimes even if it's a couple decades off, and the moment it arrives the world will rapidly change. I hope a group like them, with a focus on friendly AI, develops it before the military manages to. The military already has robots that kill people; I don't really want to know what will happen if they develop greater than human intelligence before those focused on friendly AI do.
- Zappato, on 10/11/2007, -5/+4There is one major thing that we can do to improve the world by orders of magnitude in our very lifetimes, including the real possibility of extending life tremendously, if not indefinitely - thus effectively solving the problem we call "death". It's called Friendly AI. If we don't do it, we're quite surely doomed, because someone will (rather soon) build a self-improving general intelligence which is NOT properly designed to maintain friendliness. Every dollar counts. Please suport the Singularity Institute. Don't let this opportunity to be a part of something hugely important (and beneficial - in fact, a matter of survival) pass you by!
- mwaser, on 10/11/2007, -1/+0Watch the video! The fact that we have so many movies with intelligent but dangerous machines should tell us something. The fact that the endings in those movies (where "we" win) are so contrived should tell us even more. We were able to avoid elimination by nukes (thus far) because we are the root of that problem and it's under our (intermittent) control. What happens when "we" aren't in control?
- wazzledoozle2, on 10/11/2007, -5/+4How the hell did this make front page...
- jokerthief, on 10/11/2007, -2/+1Boo on the tinyurl link. Post the real URL or I'm not clicking.
- struldbrugg, on 10/11/2007, -1/+0Pfft, go bother a real cult like the Scientologists or the Raelians rather than a bunch of secular, scientifically-minded rationalists who happen to be worried about certain future technologies and who try to do something about those worries using, yes, money. You might have a point if they ran around making predictions like "there will be a singularity on 21 December, 2012", but they don't, and you can't fault them for just saying "this may happen in the next few decades and it's worth addressing in advance".
I can see how some of the singularity claims might sound bizarre on the surface, but a lot of reasoning has gone into them. If you're interested, I recommend reading this: http://singinst.org/AIRisk.pdf - SallyMorem, on 10/11/2007, -1/+0I had trouble with the sound cutting out quite often during the video.
But the information presented is VERY important. I doubt if some wonderful programmer is going to be able to write code and create true AI on his or her own. My guess is that AI will be an emergent property, coming into being as present-day utilitarian AI software gets more and more complex, and computer systems provide environments in which they can evolve.
On producing friendly AI: If some really creative programmers can come up with ways to embed some version of Asimov's Three Laws of Robotics into present-day AI software, somewhere at the very "bottom," (whatever that may mean), I think it's possible that future generations of truly aware AIs may be guided by that, know it, and fully approve of it. At least that is my hope. I'm not a programmer, so take these comments as hopeful suggestions from an interested "civilian." - elitexero, on 10/11/2007, -2/+1The second guy has a REALLY creepishly long neck
- AUGGIEMOY, on 10/11/2007, -1/+0I'm sure in Pete's mind chicks dig him, too. (har har) Singularity's got alot of kinks to work out, but its also an incredibly widespread implications if not done correctly. But with resolving key issues, with computers that are actually smarter than we are, think of all the world catastrophe and atrocities that could be avoided. Maybe a computer can figure out how to reverse global warming, maybe fix cancer, maybe figure out world hunger.
Maybe movies dont stray far from reality. - msillitoe, on 10/11/2007, -3/+2A personal message, some of you may dig:
Aubrey de Grey, hi.
(Anamartic; the 11/780 and your II/1)
Say hello to Alice for me,
Matt
pp 0. - SallyMorem, on 10/11/2007, -1/+0One science fiction writer (I can't remember which) came up with a clever solution to the potential population problem posed by the solution to aging and death: If you take the treatment, you get kicked off the planet.
Space development, anyone? :) - keithelis, on 10/11/2007, -1/+0If anyone ever builds an artificial intelligence, SIAI is the only group in the world I would trust to be its parents, raise it to adulthood, and beyond. We have only one shot at getting the Singularity right, and even one small failure can be catastrophic. Thanks, SIAI.
-
Show 51 - 85 of 85 discussions
© Digg Inc. 2009
— Content created and posted by Digg users is