Probably the title “Dancing with Steers” strikes you as nonsensical. “What’s this about?” you ask. “A contest at the country fair in which children from the 4H Club dance the tango with their prize heifers or cowgirls dance with steers?”
The image might seem like a cartoon from Gary Larson, one reminiscent of his panel with three cows grazing and one cow commenting, “Hey, wait a minute! This is grass! We’ve been eating grass!” Yes, strange, but now I’m having difficulty getting the image of children dancing with cows out of my head; the image is a brain-worm analogous to an ear-worm song one might sing repeatedly. But that’s the point of the title: “Dancing with the Steers” is, however silly, a representation that you can imagine even though no such contest exists and cows would be hard-pressed to learn the foxtrot, the waltz, or some innovative modern dance moves. And yet, you have no problem envisioning the strange contest with those and other dance moves, with children or adults, with the handicapped or the alien, with Casper the Ghost or Mickey and Minnie.
That’s where we are as conscious beings, capable of what we might term awareness of the strange, the unlikely, even the so-called impossible. That’s what separates us from Artificial Intelligence, and in a contradiction to what I just said about our ability to think the unthinkable, I find it hard to imagine that AI will achieve a human level of existence regardless of new developments in computer sciences and processing methodologies based on the human brain. Brains have had billions of years to develop minds, and much in minds isn’t even real.
In an attempt to circumvent the problems associated with von Neumann computing systems, IBM researchers have designed a new computing architecture, one supposedly imitative of your brain.* Based on how three rare metals undergo phase changes from solid to liquid, the system works about 200 times faster than a traditional design. That’s good, but not good enough. It might be promising as a new way of computing, but it’s a long way from those superior humanoids of science fiction, such as those in RoboCop, Blade Runner, Universal Soldier, A. I. ,Star Trek , and, well, you can probably think of others.
There is the famous Alan Turing Test, of course. If the computer can fool me, it has supposedly achieved the level of development human brains have achieved. Of course, I have been fooled by robocalls as evidenced by my responding to such a caller asking if I want a new home security system (“No,” by the way). So, maybe you should be the participant in the Turing Test. Then there’s the famous Chinese Room test of John R. Searle. As Searle has said, “In my view, the greatest single failure—or class of failures, I should say—in traditional Artificial Intelligence is the failure to come to terms with the nonrepresentational character of the background.”** It’s that background stuff that causes the problem. A computer might be able to write a message in Chinese, but how do we prove that the computer actually understands Chinese. Your smart phone does some pretty amazing things, but you would have difficulty proving that Siri does little more than offer a set of probable responses with a syntax and grammar built idiosyncratically into the system. Siri isn’t a demonstrably provable being with a mind. Searle’s background includes “commonsense knowledge, practices, behavior, ways of doing things, know-how, and such preintentional capacities, both cultural and biological. There is a theoretical obstacle to programming the background capacities: they are not themselves representational. So, there is no way that their essential features can be captured in representations, but by definition, traditional Artificial Intelligence uses representations” (Baumgartner, 206, 207).
Searle and others have written extensively on the Chinese Room experiment. It’s been the subject of philosophy, psychology, and computing classes in universities and in professional journals. It centers on the problem of mind-and-body, and that’s where we get stuck eating the same grass and occasionally, like Larson’s cow, lifting our heads in realization that we keep asking the same question without realizing that the very asking isn’t something artificial intelligence would do.
I don’t know how you intend to spend the rest of your day, but I’m going to an imaginary county fair to watch some children dance with heifers or cowgirls dance with steers.
*Hiroto Kase, Ryota Negishi, Michiharu Arifuku, Noriko Kiyoyanagi, Yoshihiro Kobayashi. Biosensor response from target molecules with inhomogeneous charge localization. Journal of Applied Physics , 2018; 124 (6): 064502 DOI: 10.1063/1.5036538
** Baumgartner, Peter and Sabine Payr, Eds. Speaking Minds : Interviews with Twenty Eminent Cognitive Scientists , Princeton University Press, 1995. “John R. Searle, Ontology Is the Question,” p. 207.
The image might seem like a cartoon from Gary Larson, one reminiscent of his panel with three cows grazing and one cow commenting, “Hey, wait a minute! This is grass! We’ve been eating grass!” Yes, strange, but now I’m having difficulty getting the image of children dancing with cows out of my head; the image is a brain-worm analogous to an ear-worm song one might sing repeatedly. But that’s the point of the title: “Dancing with the Steers” is, however silly, a representation that you can imagine even though no such contest exists and cows would be hard-pressed to learn the foxtrot, the waltz, or some innovative modern dance moves. And yet, you have no problem envisioning the strange contest with those and other dance moves, with children or adults, with the handicapped or the alien, with Casper the Ghost or Mickey and Minnie.
That’s where we are as conscious beings, capable of what we might term awareness of the strange, the unlikely, even the so-called impossible. That’s what separates us from Artificial Intelligence, and in a contradiction to what I just said about our ability to think the unthinkable, I find it hard to imagine that AI will achieve a human level of existence regardless of new developments in computer sciences and processing methodologies based on the human brain. Brains have had billions of years to develop minds, and much in minds isn’t even real.
In an attempt to circumvent the problems associated with von Neumann computing systems, IBM researchers have designed a new computing architecture, one supposedly imitative of your brain.* Based on how three rare metals undergo phase changes from solid to liquid, the system works about 200 times faster than a traditional design. That’s good, but not good enough. It might be promising as a new way of computing, but it’s a long way from those superior humanoids of science fiction, such as those in RoboCop, Blade Runner, Universal Soldier, A. I. ,Star Trek , and, well, you can probably think of others.
There is the famous Alan Turing Test, of course. If the computer can fool me, it has supposedly achieved the level of development human brains have achieved. Of course, I have been fooled by robocalls as evidenced by my responding to such a caller asking if I want a new home security system (“No,” by the way). So, maybe you should be the participant in the Turing Test. Then there’s the famous Chinese Room test of John R. Searle. As Searle has said, “In my view, the greatest single failure—or class of failures, I should say—in traditional Artificial Intelligence is the failure to come to terms with the nonrepresentational character of the background.”** It’s that background stuff that causes the problem. A computer might be able to write a message in Chinese, but how do we prove that the computer actually understands Chinese. Your smart phone does some pretty amazing things, but you would have difficulty proving that Siri does little more than offer a set of probable responses with a syntax and grammar built idiosyncratically into the system. Siri isn’t a demonstrably provable being with a mind. Searle’s background includes “commonsense knowledge, practices, behavior, ways of doing things, know-how, and such preintentional capacities, both cultural and biological. There is a theoretical obstacle to programming the background capacities: they are not themselves representational. So, there is no way that their essential features can be captured in representations, but by definition, traditional Artificial Intelligence uses representations” (Baumgartner, 206, 207).
Searle and others have written extensively on the Chinese Room experiment. It’s been the subject of philosophy, psychology, and computing classes in universities and in professional journals. It centers on the problem of mind-and-body, and that’s where we get stuck eating the same grass and occasionally, like Larson’s cow, lifting our heads in realization that we keep asking the same question without realizing that the very asking isn’t something artificial intelligence would do.
I don’t know how you intend to spend the rest of your day, but I’m going to an imaginary county fair to watch some children dance with heifers or cowgirls dance with steers.
*Hiroto Kase, Ryota Negishi, Michiharu Arifuku, Noriko Kiyoyanagi, Yoshihiro Kobayashi. Biosensor response from target molecules with inhomogeneous charge localization. Journal of Applied Physics , 2018; 124 (6): 064502 DOI: 10.1063/1.5036538
** Baumgartner, Peter and Sabine Payr, Eds. Speaking Minds : Interviews with Twenty Eminent Cognitive Scientists , Princeton University Press, 1995. “John R. Searle, Ontology Is the Question,” p. 207.
RSS Feed