Comments on: Early retirement — how soon? http://www.foresight.org/nanodot/?p=2994 examining transformative technology Wed, 03 Apr 2013 18:23:47 +0000 hourly 1 http://wordpress.org/?v=3.0.4 By: http://www.foresight.org/nanodot/?p=2994#comment-828736 Tue, 31 Mar 2009 16:42:06 +0000 http://www.foresight.org/nanodot/?p=2994#comment-828736 I'm more interested in A.I. applications to daily life. For example, does the U.S. Constitution prohibit an A.I. from running for Congress or the Senate? Dreaming of an A.I. run government with the corruption software turned off and programmed to only follow the Constitution. Maybe it would be simpler to have A.I. be judges on the Supreme Court. I'm assuming A.I. wouldn't be able to ignore reality and claim the Constitution is a living and breathing document. I’m more interested in A.I. applications to daily life. For example, does the U.S. Constitution prohibit an A.I. from running for Congress or the Senate? Dreaming of an A.I. run government with the corruption software turned off and programmed to only follow the Constitution. Maybe it would be simpler to have A.I. be judges on the Supreme Court. I’m assuming A.I. wouldn’t be able to ignore reality and claim the Constitution is a living and breathing document.

]]> By: JamesG http://www.foresight.org/nanodot/?p=2994#comment-828119 JamesG Mon, 30 Mar 2009 09:56:13 +0000 http://www.foresight.org/nanodot/?p=2994#comment-828119 Communism and free riders are not liked in America, because someone else has to pay for them. With AI and nanorobotics, nobody will have to make up the slack as people retire early. Big difference. And I don't think politics or people's prejudice against advanced technologies is going to make any difference, it didn't make any difference with the TV, or the internet or social websites, and AI and nanorobotics will be far more desirable. Anyway, I don't think it is going to take that long, supercomputers with AI should be here with 2 or 3 years, people are going to be blown away by these and science will become a majorly funded field then, imo. Communism and free riders are not liked in America, because someone else has to pay for them. With AI and nanorobotics, nobody will have to make up the slack as people retire early. Big difference. And I don’t think politics or people’s prejudice against advanced technologies is going to make any difference, it didn’t make any difference with the TV, or the internet or social websites, and AI and nanorobotics will be far more desirable.

Anyway, I don’t think it is going to take that long, supercomputers with AI should be here with 2 or 3 years, people are going to be blown away by these and science will become a majorly funded field then, imo.

]]> By: http://www.foresight.org/nanodot/?p=2994#comment-827144 Fri, 27 Mar 2009 06:47:52 +0000 http://www.foresight.org/nanodot/?p=2994#comment-827144 Tech aside, I live in a 100 year old building. Just because we can do something doesn't mean we will... at least not right away. There’s sunk costs to ponder. There's recoupment time, and testing and cultural acceptance that needs to be considered. Out sourcing has slowed because its politically unpopular- that being said we certainly have the ability to outsource far, far more jobs than we have. If AI hits in 2020, will it be prevalent by 2040? Maybe. If it hits 2035, I doubt it'll be big by 2040. There's also political questions to consider. I was never clear as to whether or not this was true, but remember the rumors that you couldn't bring a G4 Mac to China because it was technically a super computer? Will other restrictions be placed on AI programs? Probably. Will they be fought tooth and nail by trade unions and worker’s advocates? Probably. Will the religious and socially conservative groups try to ban them from all walks of life… I’d wager the chances are better than average. Trying to determine when we’ll be able to achieve human level intelligence is complex enough, adding to it when we will have the political will (and economic stability) is another matter entirely. I’m openly skeptical about James Albus’s economic ideas (he’s not an economist) and a little worried about aspects of the picture Robin Hanson paints... I think a molecular assembler will be a prerequisite for the singularity, not a by product. Maybe I’m projecting my love of science fiction here, but when I talk about molecular assemblers, I mean the machine that can make anything, and I’m including in this food. If food and shelter become so cheap that employment is only a means for acquiring luxuries than sure, a massive rollout of robots and an intelligence explosion will be fine and not create vast amounts of social upheaval. But if it’s the other way around, given how Americans loath Free riders and anything that could be spun to have the faintest whiff of communism to it, don’t hold your breath on AI having an easy path to market. It would fundamentally change the political and economic landscape too much too quickly for the tastes of the average powerbrokers. As a group we spend too much time congratulating ourselves for having accomplished steps 1-2 and wondering when step 10 will arrive. We should take a harder look at steps 3-9, not all of which are technical in nature. -MissedCall Tech aside, I live in a 100 year old building. Just because we can do something doesn’t mean we will… at least not right away. There’s sunk costs to ponder. There’s recoupment time, and testing and cultural acceptance that needs to be considered. Out sourcing has slowed because its politically unpopular- that being said we certainly have the ability to outsource far, far more jobs than we have.

If AI hits in 2020, will it be prevalent by 2040? Maybe. If it hits 2035, I doubt it’ll be big by 2040. There’s also political questions to consider. I was never clear as to whether or not this was true, but remember the rumors that you couldn’t bring a G4 Mac to China because it was technically a super computer? Will other restrictions be placed on AI programs? Probably. Will they be fought tooth and nail by trade unions and worker’s advocates? Probably. Will the religious and socially conservative groups try to ban them from all walks of life… I’d wager the chances are better than average.

Trying to determine when we’ll be able to achieve human level intelligence is complex enough, adding to it when we will have the political will (and economic stability) is another matter entirely. I’m openly skeptical about James Albus’s economic ideas (he’s not an economist) and a little worried about aspects of the picture Robin Hanson paints…

I think a molecular assembler will be a prerequisite for the singularity, not a by product. Maybe I’m projecting my love of science fiction here, but when I talk about molecular assemblers, I mean the machine that can make anything, and I’m including in this food. If food and shelter become so cheap that employment is only a means for acquiring luxuries than sure, a massive rollout of robots and an intelligence explosion will be fine and not create vast amounts of social upheaval. But if it’s the other way around, given how Americans loath Free riders and anything that could be spun to have the faintest whiff of communism to it, don’t hold your breath on AI having an easy path to market. It would fundamentally change the political and economic landscape too much too quickly for the tastes of the average powerbrokers.

As a group we spend too much time congratulating ourselves for having accomplished steps 1-2 and wondering when step 10 will arrive. We should take a harder look at steps 3-9, not all of which are technical in nature.

-MissedCall

]]> By: http://www.foresight.org/nanodot/?p=2994#comment-827112 Fri, 27 Mar 2009 05:39:08 +0000 http://www.foresight.org/nanodot/?p=2994#comment-827112 I really think that there is too much emphasis on the emulation and not enough on results. We could augment AI with crowd sourcing. Narrow AI, encounters a problem, determines if it can solve it, if yes no problem, if no, then it appeals to a crowd or mechanical turk or the like. There's also the possiblity that BMI/BCI moves faster than AI and we augment our computers/robots with actual synthetic brains. I really think that there is too much emphasis on the emulation and not enough on results. We could augment AI with crowd sourcing. Narrow AI, encounters a problem, determines if it can solve it, if yes no problem, if no, then it appeals to a crowd or mechanical turk or the like.

There’s also the possiblity that BMI/BCI moves faster than AI and we augment our computers/robots with actual synthetic brains.

]]> By: http://www.foresight.org/nanodot/?p=2994#comment-826699 Thu, 26 Mar 2009 09:49:49 +0000 http://www.foresight.org/nanodot/?p=2994#comment-826699 ??? ???

]]> By: http://www.foresight.org/nanodot/?p=2994#comment-826698 Thu, 26 Mar 2009 09:49:19 +0000 http://www.foresight.org/nanodot/?p=2994#comment-826698 How much will a AI be ->worthcosts How much will a AI be ->worthcosts

]]>