2 of 7
2
Kurzweil Responds: Don’t Underestimate the Singularity
Posted: 05 March 2012 01:31 PM   [ Ignore ]   [ # 16 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  3121
Joined  2008-04-07
FreeInKy - 05 March 2012 01:25 PM

The debate about human “understanding” versus the way AI manipulates symbols without “understanding” has always fascinated me. For me, the real question is: what does it mean to “understand”? And as Turing seems to have thought, if the end result is indistinguishable, what does it matter?

I don’t think it does matter. We just seem to “care” more about what we think. We have always placed ourselves front and center in the cosmos - and we continue to discover we aren’t the front or center.

 Signature 

Turn off Fox News - Bad News For America
(Atheists are myth understood)

Profile
 
 
Posted: 05 March 2012 01:31 PM   [ Ignore ]   [ # 17 ]
Administrator
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  15435
Joined  2006-02-14
FreeInKy - 05 March 2012 01:25 PM

The debate about human “understanding” versus the way AI manipulates symbols without “understanding” has always fascinated me. For me, the real question is: what does it mean to “understand”? And as Turing seems to have thought, if the end result is indistinguishable, what does it matter?

Right. Seems to me that “understanding” is a sort of backdoor attempt to push substance dualism, though honestly how a separate mental stuff is supposed to answer the question I have no idea, either.

But the basic question, as you point out, is what is it supposed to be to understand something, above and beyond being able to use it or define it properly. Is “understanding” supposed to be some particular kind of quale? Like a special kind of shiver? But clearly we can get that shiver even when we don’t understand the thing at all. So ... forgive me but I don’t understand it.

 Signature 

Doug

-:- -:—:- -:—:- -:—:- -:—:- -:—:-

El sueño de la razón produce monstruos

Profile
 
 
Posted: 05 March 2012 04:40 PM   [ Ignore ]   [ # 18 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  447
Joined  2012-02-02
traveler - 05 March 2012 12:44 PM
Coldheart Tucker - 05 March 2012 10:43 AM

Traveler, not at all.  Kurtzweil thinks that by 2045 3D printers will be capable of manipulating things at the subatomic level.  Even if they can only work at the molecular level, that’s good enough to build a nuke.

But do you believe that? Or is it outlandish in your opinion.

The ultimate goal of the folks building things like MakerBot is to have a 3D printer capable of replicating itself (and that’s making every component, not assembling them from parts supplied by a manufacturer).  Twenty five years ago, when 3D printers were introduced, all they could do was make parts out of a few kinds of plastics.  Now, you can use them to print bone, chicken hearts, your dinner, and host of other materials (foundries are using them to print out their sand molds, for example).

According to the experts in the 3D printing industry, we’re essentially at the point where the Apple ][ was introduced, if you want to compare 3D printers to PCs.  We went from the first models costing to close to a million dollars in the late 80s, to models with identical capabilities costing roughly $1K, a quarter of a century later.  The first electronic computers were built in the mid-40s at millions of dollars, while the Apple ][ debuted roughly 30 years later at a cost of a few thousand.  So 3D printers are advancing much faster than computers originally did.  Having a 3D printer capable of building things up from the atomic level seems entirely possible to me, and potentially could show up much faster than Kurzweil predicts.

As for someone being able to use that printer to build a nuke, even if they don’t have a source of uranium or plutonium, it isn’t implausible at all.  This article, which predates the concept of nanotech by a few years, describes how a small collection of self-replicating robots could be dumped anywhere on dry land and quickly begin building a massive solar farm in short order.  Having them, or nanobots build a nuke wouldn’t be that difficult.

Another issue I take with Kurzweil is that he fails to recognize that this technology will mean the end of most commerce.  He thinks that people will buy plans for a Ferrari from Ferrari, download them into a 3D printer and have it print out a car.  But why should I pay for those plans when I can find open source plans for a knock-off Ferrari on the internet for free?  (Heck, I can have my army of nanobots crawl all over a real Ferrari and have them duplicate it for me, if I really wanted.)

Kurzweil’s “solution” to people wanting to build nukes and bioweapons is to not put certain information on the internet and to have people report for an annual brain scan by the government to verify that they’re not working on WMDs.  Not putting it on the internet is a laughable suggestion, as recent events involving the genetically engineered strain of bird flu have shown, and allwoing the government to scan the brains of people means handing over all power to the government.  That last part might sound extreme, but think about it: Power corrupts, and in those brain scans will be the data which shows what a politician would need to say to get enough people to vote for him in large enough numbers to win the election.  There’s also all the dirt that’s hidden in your opponents head.  Its simply too tempting a target to left alone by the unscrupulous.

 Signature 

“There will come a time when it isn’t ‘They’re spying on me through my phone’ anymore. Eventually, it will be ‘My phone is spying on me’.” ― Philip K. Dick

The Atheist in the Trailer Park

Profile
 
 
Posted: 05 March 2012 05:56 PM   [ Ignore ]   [ # 19 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  2425
Joined  2007-07-05
FreeInKy - 05 March 2012 01:25 PM

The debate about human “understanding” versus the way AI manipulates symbols without “understanding” has always fascinated me. For me, the real question is: what does it mean to “understand”? And as Turing seems to have thought, if the end result is indistinguishable, what does it matter?

Actually this is obvious in human behavior.  Plenty of people pretend they understand things when in actuality they have only memorised some limited part of a subject.  Like watching people troubleshoot and try to repair electronic equipment.  Lots of equipment has common problems and it is just a matter of having seen the problem before.  But when the problem is unusual or the worst case is when there are two things wrong which both create the same symptom then understanding the electronics and tracking things down is required.  The machines don’t care about words and symbols.  It is following the physics.

The Turing Test doesn’t work in getting a broken machine running.

psik

 Signature 

Fiziks is Fundamental

Profile
 
 
Posted: 05 March 2012 06:55 PM   [ Ignore ]   [ # 20 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  1201
Joined  2009-05-10
psikeyhackr - 05 March 2012 08:41 AM

The problem is that von Neumann devices are SYMBOL MANIPULATING MACHINES.  Human brains can manipulate symbols but we can also understand what the symbols mean.  Where is the computer that can reliably recognise pictures of animals?  Like tell a flock of sheep from a herd of cows or a gaggle of geese?

Computer vision is not the same as computational cognition. They are completely different problem domains. You’re looking too high-level. Our minds can manipulate symbols algorithmically, yes, but so can individual neurons. Neurons work with action potentials, which represent information about the world (for example, information from retinal cells), which means they are symbolic. At the lowest level, computers also work with voltages, 0s and 1s, which also represent information. These are the symbols they are talking about. They’re not talking about symbols with complex, cultural meanings that require years of experience to be fully understood. Computers are symbol manipulating machines just like our brains (neurons) are symbol manipulating machines.

But since you bring it up, what do you think it means to understand a (high-level) symbol? And why don’t you think computers could ever do it?

 Signature 

“What people do is they confuse cynicism with skepticism. Cynicism is ‘you can’t change anything, everything sucks, there’s no point to anything.’ Skepticism is, ‘well, I’m not so sure.’” -Bill Nye

Profile
 
 
Posted: 06 March 2012 08:32 AM   [ Ignore ]   [ # 21 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  3121
Joined  2008-04-07
Coldheart Tucker - 05 March 2012 04:40 PM
traveler - 05 March 2012 12:44 PM
Coldheart Tucker - 05 March 2012 10:43 AM

Traveler, not at all.  Kurtzweil thinks that by 2045 3D printers will be capable of manipulating things at the subatomic level.  Even if they can only work at the molecular level, that’s good enough to build a nuke.

But do you believe that? Or is it outlandish in your opinion.

The ultimate goal of the folks building things like MakerBot is to have a 3D printer capable of replicating itself (and that’s making every component, not assembling them from parts supplied by a manufacturer).  Twenty five years ago, when 3D printers were introduced, all they could do was make parts out of a few kinds of plastics.  Now, you can use them to print bone, chicken hearts, your dinner, and host of other materials (foundries are using them to print out their sand molds, for example).

According to the experts in the 3D printing industry, we’re essentially at the point where the Apple ][ was introduced, if you want to compare 3D printers to PCs.  We went from the first models costing to close to a million dollars in the late 80s, to models with identical capabilities costing roughly $1K, a quarter of a century later.  The first electronic computers were built in the mid-40s at millions of dollars, while the Apple ][ debuted roughly 30 years later at a cost of a few thousand.  So 3D printers are advancing much faster than computers originally did.  Having a 3D printer capable of building things up from the atomic level seems entirely possible to me, and potentially could show up much faster than Kurzweil predicts.

As for someone being able to use that printer to build a nuke, even if they don’t have a source of uranium or plutonium, it isn’t implausible at all.  This article, which predates the concept of nanotech by a few years, describes how a small collection of self-replicating robots could be dumped anywhere on dry land and quickly begin building a massive solar farm in short order.  Having them, or nanobots build a nuke wouldn’t be that difficult.

Another issue I take with Kurzweil is that he fails to recognize that this technology will mean the end of most commerce.  He thinks that people will buy plans for a Ferrari from Ferrari, download them into a 3D printer and have it print out a car.  But why should I pay for those plans when I can find open source plans for a knock-off Ferrari on the internet for free?  (Heck, I can have my army of nanobots crawl all over a real Ferrari and have them duplicate it for me, if I really wanted.)

Kurzweil’s “solution” to people wanting to build nukes and bioweapons is to not put certain information on the internet and to have people report for an annual brain scan by the government to verify that they’re not working on WMDs.  Not putting it on the internet is a laughable suggestion, as recent events involving the genetically engineered strain of bird flu have shown, and allwoing the government to scan the brains of people means handing over all power to the government.  That last part might sound extreme, but think about it: Power corrupts, and in those brain scans will be the data which shows what a politician would need to say to get enough people to vote for him in large enough numbers to win the election.  There’s also all the dirt that’s hidden in your opponents head.  Its simply too tempting a target to left alone by the unscrupulous.

Excellent and thoughtful reply CT. At this point, I do have to say that it still seems overly optimistic to go from 3D printers of essentially homogeneous structures to Star Trek replicators. Your analogy of the growth of the PC is a good one that does make me wonder what 3D printing might become. The PC did overcome severe limitations (primarily speed, parallelism, and memory) and perhaps 3D printing can overcome its current limitations. What makes you believe that the technology can work at the molecular level? Do you have links to anything on that? It’s quite different from squirting syringes and layering a monolith.

I like your thinking about the openness of things like Ferrari plans. But the scale of a Ferrari (in terms of expense of materials and shear size) is much different from the average user printing shower curtain rings, no? BTW, I don’t believe you really could have an army of nanobots crawl all over a real Ferrari and duplicate it for you, no matter how hard you really want it. But I do like your optimistic vision and I am cheering for you to be right and for me to be wrong. It is a pleasure to read your thoughts.

 Signature 

Turn off Fox News - Bad News For America
(Atheists are myth understood)

Profile
 
 
Posted: 06 March 2012 09:35 AM   [ Ignore ]   [ # 22 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  2425
Joined  2007-07-05
domokato - 05 March 2012 06:55 PM
psikeyhackr - 05 March 2012 08:41 AM

The problem is that von Neumann devices are SYMBOL MANIPULATING MACHINES.  Human brains can manipulate symbols but we can also understand what the symbols mean.  Where is the computer that can reliably recognise pictures of animals?  Like tell a flock of sheep from a herd of cows or a gaggle of geese?

Computer vision is not the same as computational cognition.

Being able to explain what is seen is a demonstration of cognition.  If that explanation cannot be given then cognition cannot be proven.

I ran across an experiment done on blind people long ago.  They were asked how many walls their rooms had.  Many of them said three.  It may have been most of them, I’m not sure.  But how much of what most people KNOW is derived from what they are told by people who can see?

Cognition cannot occur without comprehending three dimensional space.  How many objects do not have three dimensions and and either move or are moved in 3-D space.  Children may not know about or think in terms of X, Y and Z but they comprehend 3-D space before they even know the symbols and the math.  We are putting the symbols into the computers then trying to explain what they mean. 

Computer control systems are just reacting to stimulus according to their programming.  They understand nothing.

psik

 Signature 

Fiziks is Fundamental

Profile
 
 
Posted: 06 March 2012 02:47 PM   [ Ignore ]   [ # 23 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  1201
Joined  2009-05-10

Annnnd…..what does it mean to understand something?

And how are brains not also just reacting to stimulus according to their programming (aka neuronal wiring)?

 Signature 

“What people do is they confuse cynicism with skepticism. Cynicism is ‘you can’t change anything, everything sucks, there’s no point to anything.’ Skepticism is, ‘well, I’m not so sure.’” -Bill Nye

Profile
 
 
Posted: 06 March 2012 06:50 PM   [ Ignore ]   [ # 24 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  2425
Joined  2007-07-05
domokato - 06 March 2012 02:47 PM

Annnnd…..what does it mean to understand something?

And how are brains not also just reacting to stimulus according to their programming (aka neuronal wiring)?

Does the definition of “shoe” in a dictionary say that a shoe is a 3-dimensional object?  Does it say that shoes wear out?

Does a computer that does not know that shoes wear out understand shoes?

shoe (sh) n.
1. A durable covering for the human foot, made of leather or similar material with a rigid sole and heel, usually extending no higher than the ankle.
2. A horseshoe.
3. A part or device that is located at the base of something or that functions as a protective covering, as:
a. A strip of metal fitted onto the bottom of a sled runner.
b. The base for the supports of the superstructure of a bridge.
c. The ferrule on the end of a cane.
d. The casing of a pneumatic tire.
4. A device that retards or stops the motion of an object, as the part of a brake that presses against the wheel or drum.
5. The sliding contact plate on an electric train or streetcar that conducts electricity from the third rail.
6. A chute, as for conveying grain from a hopper.
7. Games A case from which playing cards are dealt one at a time.
8. shoes Informal
a. Position; status: You would understand my decision if you put yourself in my shoes.
b. Plight: I wouldn’t want to be in her shoes.

That definition provides no info about all of the different types and colors of shoes so how would a computer understand them without vision?  And even with vision how would it understand the problems of shoes that don’t fit well?  There are so many aspects to understanding.

I suppose the word durable might imply that they wear out if the INTELLIGENT ENTITY can consider that there might be a limit to durability but understanding occurs in levels.  AS soon as you understand one level of a subject that usually raises another level of questions to be answered.  So the entity must decide where to stop among all of the things to be interested in.

psik

 Signature 

Fiziks is Fundamental

Profile
 
 
Posted: 06 March 2012 07:17 PM   [ Ignore ]   [ # 25 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  6162
Joined  2009-02-26

psik,

Another layman’s (me) observation.

Are you saying that a newborn child understands 3D space? Why then does it take several tries for a baby to touch its hands? Why does it need to discover its hands and look at it with complete wondermend as to what this “thing” is and how it works. i have watched my child discover its hands and was fascinated by her brain trying to learn amd understand. Why does it take a child months to learn how to walk without bumping into everything? The brain learns as it experiences the world around it.
All the attributes of a shoe (your example) are learned by the brain. IMO, that means it can be learned by a computer.

I will stipulate that the trick is to find a program (logarithm) for “curiosity and experimentation to find out how and why it works” for a newborn supercomputer. The rest is accumulated data (experience).

[ Edited: 06 March 2012 07:21 PM by Write4U ]
 Signature 

Art is the creation of that which evokes an emotional response, leading to thoughts of the noblest kind.
W4U

Profile
 
 
Posted: 06 March 2012 07:43 PM   [ Ignore ]   [ # 26 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  2715
Joined  2011-04-24
Coldheart Tucker - 04 March 2012 11:16 PM

Having read Kurzweil’s The Singularity is Near, I can firmly state that he’s probably only about half right in what he thinks.  First of all, he thinks that Moore’s Law is involute, and that it’ll never accelerate, nor decelerate.  The reality is that technological progress is unpredictable, and there have been several announcements in recent weeks which indicate that it might be possible to completely leap frog Moore’s Law and jump to near ultimate processing power chips in just a few years, rather than taking until 2045 or so, as Kurzweil predicts.

Kurtzweil also thinks that technology will completely solve all of our problems, while failing to realistically address the downsides of the technology.  For example, Kurtzweil estimates that by 2045 you’ll be able to buy a laptop for about $1K which will have a processor so fast that it can essentially accomplish every mental computation that humanity has done in the whole of its existence, in just a few seconds.  Certainly possible and likely, IMHO.  Couple this with 3D printers that have nearly the same capabilities as a Star Trek replicator, and there’s the serious potential for disaster.  If I’m a terrorist wanting to make a nuke or a bioweapon, I no longer need millions of dollars, advance degrees, or connections to people who have the skills and resources that I lack.  I just ask my laptop (I might have to “jailbreak it” to unleash this capability, but probably not) to figure it out, load the plans into my 3D printer, and I’m Dr. Evil faster than you can say “Bob’s your uncle.”

There’s a lot more, but I don’t have time to go over it.  I will note that Kurtzweil believes (though he doesn’t come right out and admit it, preferring to drop coy hints) that the universe deliberately created humanity (and we’re the only intelligent life in the universe) so that we can become Borg like beings who assimilate the universe ala V’Ger in Star Trek the Motion Picture.  Again, I don’t think that he’s completely full of feces, but I’d take anything he says with a grain of salt.

I have been interested in transhumanism for a few years and in IMO, Kurzweil seems to have an almost religious zeal when describing the singularity; some other transhumanists and scientists have taken issue with his predictions, and I agree with them. That said, it’s still fun to think about.

 Signature 

Raise your glass if you’re wrong…. in all the right ways.

Profile
 
 
Posted: 06 March 2012 09:57 PM   [ Ignore ]   [ # 27 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  447
Joined  2012-02-02
traveler - 06 March 2012 08:32 AM

Excellent and thoughtful reply CT. At this point, I do have to say that it still seems overly optimistic to go from 3D printers of essentially homogeneous structures to Star Trek replicators. Your analogy of the growth of the PC is a good one that does make me wonder what 3D printing might become. The PC did overcome severe limitations (primarily speed, parallelism, and memory) and perhaps 3D printing can overcome its current limitations. What makes you believe that the technology can work at the molecular level? Do you have links to anything on that? It’s quite different from squirting syringes and layering a monolith.

I like your thinking about the openness of things like Ferrari plans. But the scale of a Ferrari (in terms of expense of materials and shear size) is much different from the average user printing shower curtain rings, no? BTW, I don’t believe you really could have an army of nanobots crawl all over a real Ferrari and duplicate it for you, no matter how hard you really want it. But I do like your optimistic vision and I am cheering for you to be right and for me to be wrong. It is a pleasure to read your thoughts.

What makes me believe in 3D printing at the molecular level is partly in Kurzweil’s writings on that specific subject (he lays out the mathematics of how it will work), but also in personal experience in the field of manufacturing.  Being able to build things up from the molecular level just opens up huge possibilities in not only what can be made, but how it can be made.  Vast amounts of wasted material can be eliminated, parts can have aspects of their material tuned at a precise level, rather than at the coarse (and less than ideal) level we currently have.  This will enable longer life for products made in this manner, as well as new types of products we can’t imagine.

As we get closer to that technology being possible, expect to see more and more organizations (though its possible by this point that you’ll not see corporations as they’ll have ceased to exist by that point) frantically pursuing the technology.

The nanobot armies will be rather common place, simply because by the time we get to the point that such things can be built, the cost to do so will be almost nil.  There are parallels in the electronics industry.  By the late 90s the cost of producing a cellphone was practically nothing, the economies of scale had gotten so large.  Smartphones will probably reach that point in anotheer couple of years.  Nanotech will be no different.

 Signature 

“There will come a time when it isn’t ‘They’re spying on me through my phone’ anymore. Eventually, it will be ‘My phone is spying on me’.” ― Philip K. Dick

The Atheist in the Trailer Park

Profile
 
 
Posted: 07 March 2012 07:29 AM   [ Ignore ]   [ # 28 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  3121
Joined  2008-04-07

That sounds pretty ambitious for a 33 year schedule. If you and Kurzweil are correct, I might just live long enough to see it.
You say organizations will replace corporations. How will these organizations be different from corporations in your opinion? And why do you believe that corporations will meet their demise in just 3 decades?

 Signature 

Turn off Fox News - Bad News For America
(Atheists are myth understood)

Profile
 
 
Posted: 07 March 2012 07:47 AM   [ Ignore ]   [ # 29 ]
Administrator
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  15435
Joined  2006-02-14
mid atlantic - 06 March 2012 07:43 PM

I have been interested in transhumanism for a few years and in IMO, Kurzweil seems to have an almost religious zeal when describing the singularity; some other transhumanists and scientists have taken issue with his predictions, and I agree with them. That said, it’s still fun to think about.

Yeah, I mean Kurzweil is a smart guy and all, but with this singularity stuff he’s basically a crank. Worth noting that he’s also involved in all kinds of quack self-medication: he claims to take 150 pills a day ...

 Signature 

Doug

-:- -:—:- -:—:- -:—:- -:—:- -:—:-

El sueño de la razón produce monstruos

Profile
 
 
Posted: 07 March 2012 08:38 AM   [ Ignore ]   [ # 30 ]
Sr. Member
Avatar
RankRankRankRankRankRankRankRankRankRank
Total Posts:  3121
Joined  2008-04-07
dougsmith - 07 March 2012 07:47 AM
mid atlantic - 06 March 2012 07:43 PM

I have been interested in transhumanism for a few years and in IMO, Kurzweil seems to have an almost religious zeal when describing the singularity; some other transhumanists and scientists have taken issue with his predictions, and I agree with them. That said, it’s still fun to think about.

Yeah, I mean Kurzweil is a smart guy and all, but with this singularity stuff he’s basically a crank. Worth noting that he’s also involved in all kinds of quack self-medication: he claims to take 150 pills a day ...

True. It’s people like Kurzweil who almost make ad hominem seem like a reasonable approach.  LOL

 Signature 

Turn off Fox News - Bad News For America
(Atheists are myth understood)

Profile
 
 
   
2 of 7
2