|
July 2007 Artificial Intelligence Is Lost in the WoodsA conscious mind will never be built out of software, argues a Yale University professor. By David Gelernter
Artificial intelligence has been obsessed with several questions from the start: Can we build a mind out of software? If not, why not? If so, what kind of mind are we talking about? A conscious mind? Or an unconscious intelligence that seems to think but experiences nothing and has no inner mental life? These questions are central to our view of computers and how far they can go, of computation and its ultimate meaning--and of the mind and how it works. They are deep questions with practical implications. AI researchers have long maintained that the mind provides good guidance as we approach subtle, tricky, or deep computing problems. Software today can cope with only a smattering of the information-processing problems that our minds handle routinely--when we recognize faces or pick elements out of large groups based on visual cues, use common sense, understand the nuances of natural language, or recognize what makes a musical cadence final or a joke funny or one movie better than another. AI offers to figure out how thought works and to make that knowledge available to software designers. It even offers to deepen our understanding of the mind itself. Questions about software and the mind are central to cognitive science and philosophy. Few problems are more far-reaching or have more implications for our fundamental view of ourselves. The current debate centers on what I'll call a "simulated conscious mind" versus a "simulated unconscious intelligence." We hope to learn whether computers make it possible to achieve one, both, or neither. I believe it is hugely unlikely, though not impossible, that a conscious mind will ever be built out of software. Even if it could be, the result (I will argue) would be fairly useless in itself. But an unconscious simulated intelligence certainly could be built out of software--and might be useful. Unfortunately, AI, cognitive science, and philosophy of mind are nowhere near knowing how to build one. They are missing the most important fact about thought: the "cognitive continuum" that connects the seemingly unconnected puzzle pieces of thinking (for example analytical thought, common sense, analogical thought, free association, creativity, hallucination). The cognitive continuum explains how all these reflect different values of one quantity or parameter that I will call "mental focus" or "concentration"--which changes over the course of a day and a lifetime. Without this cognitive continuum, AI has no comprehensive view of thought: it tends to ignore some thought modes (such as free association and dreaming), is uncertain how to integrate emotion and thought, and has made strikingly little progress in understanding analogies--which seem to underlie creativity. My case for the near-impossibility of conscious software minds resembles what others have said. But these are minority views. Most AI researchers and philosophers believe that conscious software minds are just around the corner. To use the standard term, most are "cognitivists." Only a few are "anticognitivists." I am one. In fact, I believe that the cognitivists are even wronger than their opponents usually say. But my goal is not to suggest that AI is a failure. It has merely developed a temporary blind spot. My fellow anticognitivists have knocked down cognitivism but have done little to replace it with new ideas. They've showed us what we can't achieve (conscious software intelligence) but not how we can create something less dramatic but nonetheless highly valuable: unconscious software intelligence. Once AI has refocused its efforts on the mechanisms (or algorithms) of thought, it is bound to move forward again. Until then, AI is lost in the woods. |
 |
|
||||
| |||||
Audio »
Share »
Digg this
Add to del.icio.us
Add to Reddit
Add to Facebook
Slashdot It!
Stumble It!
Add to Newsvine
Add to Connotea
Add to CiteUlike
Add to Furl
Googlize this
Add to Rojo
Add to MyWeb
Favorite
Print
E-mail
Comments
jb.design@yahoo.com.au on 06/25/2007 at 4:31 AM
1
1: Where does empathy fit into this equation? If an individual suffers from mental disorders, which inhibit they’re ability to empathize with others (psychotics), or they perceive/experience the world differently to the statues quo, are these people not human still? For instance if they do not feel pain, or are quadriplegics.
2: My final thought and I’ve had this one for a while is as such:
If we used the internet as our global database of knowledge, and then we use Google or another competent search engine to search through the WWW.
A bit of software could be asked questions on anything, and all the software would be required to perform, is a Google search on the WWW, and the web links would only have to be information scrapped and understood, within the context of the question, so if the software could not only understand the grammatical nature of the language being used, but also understand the nature and structure of the question being asked, it would only need to structure an appropriate answer, and the information used by the software to answer the question could all come from Google.
Now imaging if you were playing chess with this software it could simply search for a chess strategy from the internet, with out you knowing and use this strategy against you to beat you.
In essence it is using other people’s knowledge and life experiences from the WWW to interact with you.
You wouldn’t have a clue that you were no longer interacting with a human but in fact with a machine, especially if the nature of engagement was changed and the interactive experience you had with the software was also changed, remove the face to face element, and change the interface to a computer chat session, hence does the nature of consciousness actually fall squarely in our own abilities to experience and understand the world? I bring this up because I used to play a lot of shooting games, and developers created Bot like characters which would run around with you while you played, they would engage with you, and engage the enemy, now I knew that they were non humans because I could turn them on or off, but in cyberspace you wouldn’t be able to tell who is human and who isn’t, it becomes very blurred, so in a realm like second life, where more an more people flock to for socialising, you have the perceived notion that all those avatars running around you are in fact other conscious beings, with out the physicality’s of the real world, but what if some of them really aren’t?
It’s like a scene out of the 'Matrix', what is real and what isn’t real, does consciousness solely lie in our physical world? If such a ‘Matrix’ style interface could be built and smart pieces of software could navigate through our virtual worlds interacting with us, and we never knew it, what then? Maybe I am a bit of smart software who exists on someone’s laptop somewhere, and I’ve information scrapped all of this from Google and placed a name at the end of it to give me some form of identity, how would you ever know?
Jay
ajimenez on 06/25/2007 at 10:39 AM
14
inboulder on 06/25/2007 at 5:41 AM
10
"In conscious thinking, you experience your thoughts. Often they are accompanied by emotions or by imagined or remembered images or other sensations. A machine with a conscious (simulated) mind can feel wonderful on the first fine day of spring and grow depressed as winter sets in."
This definition of consciousness is hopeless. Nowhere did I see mention of the level of complexity of reflexive systems necessary to create consciousness, nor of internal modeling, nor of feedback, just some blustering about feelings.
""Ever felt hungry or thirsty?" "Itchy, sweaty, tickled, excited, conscience stricken?"
"Ever mourned?" "Ever rejoiced?"
No, no, no, no."
Does a chimp feel these things? A dog, a squirrel, a lizard, a bug? Consciousness is obviously not an either/or thing, it is a continuum. It's not some magic box. It's clearly possible to be a bit conscious, more conscious etc.
"Everything computers accomplish is built out of these primitive instructions."
Um, everything in our brains is built out of primitive analog feedback devices.
"Computers don't know or care what instructions they are executing. "
Your neurons don't know or care why they fire.
"And the computer is merely a machine doing what it's supposed to do"
Your brain is merely a biological machine doing what it's supposed to do.
"Cognitivists argue that sure, you know what executing low-level instructions slowly is like; but only when you do them very fast is it possible to create a new conscious mind"
Really, I've never heard a congnitivist argue that, in fact it seems perfectly straightforward that it's possible to emulate a human brain by hand, but doing those quadrillions of analog feedback operations is going to be pretty slow, but there's no reason why it's not possible.
"But we feel emotions with our whole bodies, not just our minds; and it has no body. "
Right, because paraplegics aren't conscious?
"Turing test--which is one index of the test's superficiality."
Of course it's superficial, I don't know any researchers that think passing the turing test is worth a hill of beans. Nobody cares about the turing test, I'm not sure why the article keeps harping on it.
"What gives us the right to inflict such cruel mental pain on a conscious being?"
Are you kidding me, you don't think someone would volunteer etc? These are the types of issues the author thinks are even worth discussing? These aren't serious anti-cognitivist objections at all.
"But there isn't any reason to believe it would."
The reason is, because we can think. We have a 6 billion examples of how it's done, we are getting better and better at understanding this model. Don't you think it likely at some point we'll understand it well enough to emulate it enough to get a bit of consciousness (probably with 'unconscious' AIs help) ?
Ok, I'm giving up on this article, it seems too lost in the woods to keep reading.
rfentres on 06/25/2007 at 5:12 PM
1
mir on 06/25/2007 at 9:15 AM
2
The author obviously does not understand anything about research in neuroscience, yet he says, he is 'student of mind' or something.
If he does not understand the 'difference between a child's consciousness and an adult's?' or the 'differences between child-style and adult-style thinking?', or 'why children tend to think more concretely than adults' etc. etc., he should not generalize to all people. There are many neuroscientists which understand it, there are many neuroscientists which can map the roots of 'creativity' and 'free associations' in the precise manner (in neural substrates in humans and animals). A great deal is known about human neocortex and related substructures and successful models have been made in ARTIFICIAL neural simulations reaching the same types of performance or errors.
Work in last 130 years in brain understanding suggests that brain IS information processing machine, yet very complex. We have recently discovered a great deal about brain, and every detail discovered suggest that neurons perform COMPUTATIONS and we have better and better picture how and what the neurons compute.
At some place the author writes:
"(computer would have) no idea what "add" means, what a "number" is, or what "arithmetic"
I ask the author, how he knows what "add" means,
what a "number" is, or what "arithmetic" is?
If he does not know the answer, I will help him:
He knows it from sensory experience, by forming concepts, and by generalization of concepts.
If he does not know anything about how brain works, it is unfair to stick this ignorance to all people, because there are many which know great deal about that.
After I came with my reading to 'continuum of mental states', 'focus level', 'highest-focus thought' I have quit reading...
rg on 06/25/2007 at 12:21 PM
2
No question the continuum idea is powerful. i'm very glad TR published this. Thanks TR and D Gelernter.
ajimenez on 06/25/2007 at 10:54 AM
14
At the high end, we must lose consciousness for how can we be aware of self when solving an equation. At the low end we are like our pets who expect play when we open the back door to our yard but fail to realize that today is Saturday and I have to mow the lawn. At the low end, we do not lose consciousness but it is primitive. And, in the middle, we have the conundrum of 'normal' consciousness--we are smart enough to know we exist but our thoughts are constantly muddied by the low end dancing with the high.
Thanks for the article, professor. It was a real eye-opener. And thank you TR for bringing it to our attention.
rarnold on 06/25/2007 at 12:55 PM
1
abcarterjr on 06/25/2007 at 2:59 PM
45
Auto Pilot hardware and software has not led
to the UNCLASSIFIED discovery of the origins of
human consciousness. Perhaps a study of the interactions
of the human brain electrochemistry with the
microbial colonies residing in the brain might
lead to a discovery of the evolution of human
consciousness.
Tom B on 06/25/2007 at 3:37 PM
9
http://en.wikipedia.org/wiki/Philosophical_zombie
alan2 on 06/25/2007 at 4:43 PM
1
Again, what almost everyone in this field does is try to steal second base in their arguments.
The assumption in the statement above is "He". There is an I existing prior to any such learning and how does it arise? By sensory experience, by forming concepts, and by generalization of concepts? Ha! You can't pass first base without earning it.
I love how these people argue themselves into non-existence. Scientists masquerading as philosophers have caused great damage to our ability to think rationally (and hence the world).
Free choice and, hence consciousness, cannot come from a deterministic or probablistic device.... except by blind faith and magic. There is always the programmer in these scenarios where the true free choice exists.
Explore consciousness and neuroscience so we can learn how the brain works - absolutely. Confusing a philosophic discussion for a scientific one - never.
mir on 06/26/2007 at 6:05 AM
2
Yes there is an existing prior. Cells, which can learn above mentioned.
Prior the discovery of DNA, genes, proteins folding etc., a lot of people like you and the author have believed that the living beeings need some 'life souce' in order to live, reproduce etc.
Today, after we know a great detail about machinery of life, we do not need magic any more.
The same is true for "consciousness", "continuum of mental states", "focus level" etc. Neurons are cells. There is no special magic souce on top of neurons. Majority of people understand those concepts and how they emerge in brains. If you or the author do not, I am sorry about that. Obviously, both of you have to work and learn a bit harder.
"Confusing a philosophic discussion for a scientific one - never."
It is not true. The truth is that such philosophic discussion is not going to reveal any truth about all the things ("consciousness" etc.) the author writes about. The reason is that philosophy do not have any tools to discover the truth about it. Indeed, it has very bad record of revealing the truth in this field. In last 40 years of modern neuroscience we have learned a lot more than in last 2000 years by philosophy. It is very good thing that the line of thinking of author is in minority.
Otherwise, science and technology would indeed be lost in the woods.
brunascle on 06/27/2007 at 2:15 PM
Web Developer
69
why not? how do you know that you're free choice is any different than a computer program? what do you believe exists inside the brain that cannot be created by man?
dansturner on 07/23/2007 at 3:37 PM
1
sacapiloa on 06/29/2007 at 3:46 PM
9
Jill's Jack on 05/08/2008 at 5:45 PM
1
skeptic on 06/25/2007 at 5:56 PM
1
Philosophers are the creationists of cognitive science...
monkeytypist on 06/25/2007 at 8:50 PM
2
brunascle on 06/27/2007 at 2:18 PM
Web Developer
69
which can be recreated. these quantum mechanical properties exist everywhere, even in electronic devices.
Cygnet on 08/16/2007 at 10:07 AM
1
Cultor on 06/25/2007 at 6:17 PM
7
rafael7 on 07/05/2007 at 11:51 AM
6
GaryB on 06/25/2007 at 6:23 PM
39
It is easy to write a program that records the human's play and use it to beat them. The reason is, the human's "free" choice is highly correlated with their past decisions even when they just try to go wild and play random. The problem is, there is no random number generator inside your head, all your actions are very correlated. The human can beat the machine using explicit or implicit counting strategies however.
There is, moreover a lot of evidence that your conscious experience occurs after the "free" choice and often after the initiation of action.
This leads me to believe that in a sense, consciousness is an illusion. It experiences the past not the present. IMHO (my guess): I think consciousness comes from simulation of self (necessary in any higher level robotics for planning) coupled with our exquisite machinery to simulate others (just look to your dreams). We need this machinery because of our highly structured social nature, and when the simulation of others is turned on our simulation of self, we get consciousness. Consciousness itself has high utility for, again, planning by simulation. I see no barriers and no magic to prevent us doing this in machines. Animals have consciousness (I believe) but to lessor degrees much as apes have language, but not our highly flexible re-combinable type of language.
Tom B on 06/25/2007 at 6:38 PM
9
You're right about the data, but I'm not sure your conclusion makes sense.
Cultor on 06/25/2007 at 8:35 PM
7
If I loose my memory can I be conscient anyway?
I think (who thinks?) consciousness is in the instant.
Tom B on 06/25/2007 at 8:45 PM
9
GaryB on 06/26/2007 at 12:52 PM
39
Brian H on 06/25/2007 at 11:23 PM
19
No memory, pretty hard to maintain that there's an "I" left at all. Consider the Star Trek Transporter. Sends a pattern of you elsewhere, which is used to reassemble you. But it can be tapped into, permitting one or n additional reassemblies, all of which are convinced they're you.
Of course, none is right. You died when the Transporter atomized you on the departure stage.
In fact, prove you're the same person who fell asleep in your bed last night. I'll assert, for the sake of discussion and disruption, that you died and a new self was created on waking with your memories, and hence the illusion of continuity. The same will happen tonight, and every night. The only way to be even reasonably sure of continuity is never to fall asleep. Good luck with that.
So memories define everything, but prove nothing.
GaryB on 06/26/2007 at 12:39 PM