Our age is a bit obsessed with robots, artificial intelligence, and simulations. Do we have some sort of built-in God Complex that we are driven to create in our own image?
Here’s where robot technology coupled with 3D printing and response polymers has taken us: Materials with self-actuation in response to different stimuli. Essentially, some clever humans at Caltech 3D-built a flat polymer sheet that incorporates hinges and that responds to 200 degrees Celsius by becoming a folding and unfolding pentagonal wheel that moves itself. In other words, the gizmo can roll along a surface in response to heating. The “five-sided pentagonal wheel,” called “Rollbot,” paves the way for “untethered soft robots,” shades of one of those Terminator movies.* Think of an auto-origami making itself auto-mobile, a shape-shifting inorganic actor responding to its environment.
Well, we’re not quite at the Terminator-2 stage yet, not quite making a T-1000 to battle Arnold Schwarzenegger’s older model T-800; nevertheless, we’re on verge of making robots that can physically respond to external stimuli by changing shape. And it’s not just the self-folding Rollbot that the Caltech group produced. One of the self-actuating robots they made appears to be a flat sheet until it is placed in water, whereupon it turns into a bug-like shape. It’s a real shape-shifter. When someone figures out how to add some neurocircuitry, we’re going to have competition as planetary rulers, or at least that was the fear of the late Stephen Hawking. Is there a limiting principle that will prevent robots from competing or from establishing superiority in the hierarchy of intelligence? I think there is, but I’ll name it below.
I can imagine such “soft” robots serving both art and technology. One could make sculptures that change shapes in a manner that previous, more rigid kinetic art could not do. And unlike your computer’s screen-saving fractals, such art would be three-dimensional and solid. Furniture, also. Think of the advantage there. “Honey, the door is too narrow to get the couch into the house.” “No, dear. Remember we bought a self-assembling couch from IKEA. It is a thin sheet until you wedge some Cheez Doodles into the slot where the pillows will form.”
And what about such soft devices sent into places where humans would be in danger or where changing shape would be necessary for access, such as in caves or collapsed tunnels, maybe also into reactors? Maybe the soft robots could make their way into tight quarters to seal a leak: They could travel into the tight spot. And then, acting like some surrogate for an “as seen on TV” product, they would unfold themselves to cover the leak. Maybe all our robotic creations are designed as helpers that would also decrease danger, robot soldiers, for example. But, then, what if robot soldiers turned on us as so many sci-fi stories project?
“Why,” I ask myself, “would a Caltech story about a soft robot not only catch my attention, but also keep it?” Is there a lesson to be learned here? Was Hawking right to warn us about our impending slavery to machines that think? Will our projected downfall be our own fault? Remember what Zeus said when the gods asked him to take sides during the Trojan War: “Man has only himself to blame if his miseries are worse than they ought to be.” Yeah, the biblical “Man,” Neil Armstrong’s “Man”: The word means, of course, “humanity” or “humans,” and it includes both genders, the male and female brains, the Earth-bound “Martians” and “Venusians.”
We’ve long been in the business of trying to create that which mimics what we are: Thinking, responding, and anticipating beings. Statues are nice, but they lack life and only imitate shape. Disney World’s “animatronic characters” imitate shape and movement, but move only to the extent that they are preprogrammed. Factory robots are very practical inventions; car bumpers are no longer misaligned and engines are built with finer tolerances than they were in the pre-robot assembly line.
But robots are at this time incapable of responding to unexpected stimuli on the order of human response, though very good at responding to stimuli we anticipate for them (Did Pavlov know what was to come?). Characters in computer games do things humans can’t, defying the physical laws of the “real” world; they are also two-dimensional and, for all their abilities, can’t move a feather. Artificial intelligences incarnate in a robot bodies are the product of piece-by-piece additions of sensors. At this time, we have some robots doing what we can do here or there, but not, as we, everywhere with everything at anytime. That’s that “responding and anticipating” that I just noted.
This new soft robot technology is a further step toward our making a being in our own image. In creating intelligent, responsive robots, we are like the mythical sculptor Pygmalion, whose statue Galatea came to life. And, now that I come to think of Pygmalion, maybe our goal outside utilitarian technology is to make companions-to-order, male or female “Stepford wives” that carry none of our human unpredictability. Shades of “sexbots”! (Do we create with the ultimate goal of satisfying our hormonal drives?)
As neurologist Oliver Sacks, author of the popular book The Man Who Mistook His Wife for a Hat: And Other Clinical Tales, noted, humans are a “miracle of integration.” That integration results in a sense of self; otherwise, we are merely more sophisticated robots that respond to our environments as biological computers. In our attempt to recreate ourselves in the form of robots, we fail to consider the principle of integration because it is too difficult to imitate. And there’s something else beyond the principle of integration: The Principle of Gender Difference. As Louann Brizendine explains in The Female Brain, there are in general identifiable differences in “Man” that Zeus, from his perch on Olympus, glossed over in his comment. Female brains, according to Brizendine, contain on average more neurons (11% more) devoted to language and hearing than men, and they have a slightly larger hippocampus. In an on-average relatively smaller female head, female brains also pack more densely the same number of neurons as men with an on-average larger (more voluminous) head—yes, guys, you have bigger heads, but you incorporate more “emptiness.” There’s a difference, even if an on-average difference. And that difference will probably keep us a step ahead of robot we create, even soft ones imbued with some very fast processors.
Creating a “human-like robot” is difficult, but, were we to do so, therein lies the danger that Stephen Hawking and others—many of them science fiction writers—tell us to avoid. When Rollbot can both change shape and employ AI while responding to different stimuli, we’re in at least a little trouble of our losing that place at the head of the hierarchy. Rollbot’s successors will have none of our frailties unless we, the creators, build them into their makeup.
But gender is a problem for robot creators. Toshiba Corp.’s Aiko Chihira, the Japanese-speaking and very female-looking robot that directed customers at the Bank of Tokyo Mitsubishi UFJ and the Mitsukoshi store in Nihonbashi, Tokyo, was a long way from being “human.” That Aiko Chihira looked human was simply a capitulation to our need to think “she” was human. A robot in the shape of a box could give the same information as the robot with human features and multiple human-like expressions. ** Aiko (if I may be so bold as to refer to “her” by her first name) is “humanoid” at best, more appearance than substance, with remarkably life-like features, but without “life.” “She” has no sense of self, and we cannot assume that she “thinks like a woman.”
If you look at Aiko and other robots, you see in them our limitations in making beings in our own image. Of course, the goal of robotics engineers is to make a robot that can sense its environment as we do and respond to stimuli. To do so means they would have to program robots for the unexpected vicissitudes of the real-world environment; and when the engineers fail to anticipate what the robot surrogate will encounter, the robot will fail. But notice that engineers design by stereotype. “Human-like” and “humanoid” images shape their engineering designs. But there are those subtle differences in “Man,” and those differences are embedded in the brains of two genders, making imitating what we all are as humans very difficult to mimic.
Pygmalion’s Galatea both looked and acted like a human female. That ancient Greek myth contains a number of assumptions, not the least of which is that a statue can become “woman-like” in the real world of marriage, birth, and motherhood (Galatea’s child was Paphos). In contrast Watson, the IBM Jeopardy champion (against Ken and Brad in 2011) had no human physical features, but outperformed the two greatest trivia champs.*** Watson had to deal only with language and historically recorded knowledge, and it seemed to be capable of making inferences to reach correct answers. But could Watson “feel”? Did it reach its answers in a linear, single-plane, though rapid, set of logical steps while Ken and Brad acquired their answers as aggregates that just popped into their consciousness through multi-planal thinking?
You might argue that as physicists and engineers approach quantum computing, soft robots will become more human-like. But there’s still going to be a difference. The cold machine computing on a quantum level will always be different from the hot brain computing that operates on as yet not fully understood multiple levels of interconnections with a central command that chooses what to attend to and what to ignore. Think of a computer’s processes as occurring on a single plane and a human’s processes as occurring on multiple planes, many of them intersecting. True, the computer can process faster than the brain, but that doesn’t make it more complex or capable of “seeing” what the brain sees in every situation.
Plus, humans aren’t really good at predicting on the basis of mathematical computation, so they assess differently from computers, which are good at processing probabilities. Give me an intuitive robot, and you’ll give me a more human-like robot. Is our robot-goal, then, to make an error-free human?
So, which “image” of “Man” is the image model for “Man”-like robots? The brain in general and all the physical properties associated with it, just the brain’s processing, or the processing and behaviors associated with and connected to stimuli and the brains of others?
Here’s my own “Turing Test”: Make an angry robot. Or a loving one. The motion picture I Am Mother, written by Michael Lloyd Green and Grant Sputore, puts a “loving,” caring robot in charge of rebooting the human race. Mother is even willing to act as a mother would (I won’t give a spoiler, but will recommend the film). But think about it, to make a human, you really have to make a human. All our efforts to create in our own image will fall short, regardless of what the science fiction writers tell us and the engineers promise to make.
And what about those sheets of polymers that seem to respond to the environment? Well, you, too, have polymers. Teamed up with Stuart Hameroff, Roger Penrose explained in his Shadows of the Mind that consciousness derives from superposed quantum states, each acting in its own spacetime geometry. Somehow, they propose, the human brain operates on a level of quantum gravity in tubular polymer structures that can be found inside neurons. Regardless of the somewhat unprovable nature of that explanation, one has to acknowledge that we don’t operate on purely logical grounds as a machine would. Look inside; there’s much going on, too much to keep track of and anyway, no one has yet been able to keep track of the components of the quantum microworld. That means that the replication or duplication of a thinking, responding, and anticipating being we know as “Man” is beyond our present and future technologies. Why go out of our way to train inorganic polymers to do what those tubular polymer structures in our neurons already do? Again, is it because we have a God Complex?
And so my answer to that question comes back to a theme I’ve hit on before. Remember the Adam and Eve story? If you reduce it to its essence, it’s about pride. They wanted to be like God. And we, in creating robot after robot, refining AI as much as we can, seem to have inherited that ancient human desire. “I want to be like God, and making a thinking, responding, and anticipating being is my path to apotheosis.”
Short of our creating in our own image some robot, we take on the task of shaping others, our children, as best we can to make them in our own image. But, as in creating the perfect human substitute in a robot, creating the perfect replica of us in our children is doomed to failure. In part, that lack of success lies in our thinking of individuals in terms of “Man,” that is, the “general human.” As every parent discovers, there are just too many variables to manipulate, to control, to anticipate. And if Penrose is correct, then we’ll never get a handle on the illogical nature of our quantum nature, where probability, and not predictability, rules.
Of course, the course is set. Engineers will continue to create robot after robot after robot. No one is going to change that unless our technological civilization fails. Ironically, there are those among us those who would, in fact, destroy civilization and their very hope of achieving such godlike status, and many of them—as terrorism of the last twenty years has shown—seek to destroy in the name of their deity. That self-destructiveness is part of “Man’s” potential, and its presence means that to truly create in our own image, we would have to incorporate it in our robot progeny. We have two kinds of brains (in general), unpredictability, and integration of Self that make up individuals and that make a creation of “Man” impossible to achieve no matter how much our desire to be godlike drives us in our creations.
*Caltech. Self-folding “Rollbot” paves the way for fully untethered soft robots. August 21, 2019.
https://www.caltech.edu/about/news/self-folding-rollbot-paves-the-way-for-fully-untethered-soft-robots
**Hongo, Jun. Robotic Customer Service? In This Japanese Store, That’s the Point. The Wall Street Journal online, April 16, 2015. https://blogs.wsj.com/japanrealtime/2015/04/16/toshiba-humanoid-robot-to-debut-in-tokyo-department-store/
*** https://www.youtube.com/watch?v=P18EdAKuC1U
Here’s where robot technology coupled with 3D printing and response polymers has taken us: Materials with self-actuation in response to different stimuli. Essentially, some clever humans at Caltech 3D-built a flat polymer sheet that incorporates hinges and that responds to 200 degrees Celsius by becoming a folding and unfolding pentagonal wheel that moves itself. In other words, the gizmo can roll along a surface in response to heating. The “five-sided pentagonal wheel,” called “Rollbot,” paves the way for “untethered soft robots,” shades of one of those Terminator movies.* Think of an auto-origami making itself auto-mobile, a shape-shifting inorganic actor responding to its environment.
Well, we’re not quite at the Terminator-2 stage yet, not quite making a T-1000 to battle Arnold Schwarzenegger’s older model T-800; nevertheless, we’re on verge of making robots that can physically respond to external stimuli by changing shape. And it’s not just the self-folding Rollbot that the Caltech group produced. One of the self-actuating robots they made appears to be a flat sheet until it is placed in water, whereupon it turns into a bug-like shape. It’s a real shape-shifter. When someone figures out how to add some neurocircuitry, we’re going to have competition as planetary rulers, or at least that was the fear of the late Stephen Hawking. Is there a limiting principle that will prevent robots from competing or from establishing superiority in the hierarchy of intelligence? I think there is, but I’ll name it below.
I can imagine such “soft” robots serving both art and technology. One could make sculptures that change shapes in a manner that previous, more rigid kinetic art could not do. And unlike your computer’s screen-saving fractals, such art would be three-dimensional and solid. Furniture, also. Think of the advantage there. “Honey, the door is too narrow to get the couch into the house.” “No, dear. Remember we bought a self-assembling couch from IKEA. It is a thin sheet until you wedge some Cheez Doodles into the slot where the pillows will form.”
And what about such soft devices sent into places where humans would be in danger or where changing shape would be necessary for access, such as in caves or collapsed tunnels, maybe also into reactors? Maybe the soft robots could make their way into tight quarters to seal a leak: They could travel into the tight spot. And then, acting like some surrogate for an “as seen on TV” product, they would unfold themselves to cover the leak. Maybe all our robotic creations are designed as helpers that would also decrease danger, robot soldiers, for example. But, then, what if robot soldiers turned on us as so many sci-fi stories project?
“Why,” I ask myself, “would a Caltech story about a soft robot not only catch my attention, but also keep it?” Is there a lesson to be learned here? Was Hawking right to warn us about our impending slavery to machines that think? Will our projected downfall be our own fault? Remember what Zeus said when the gods asked him to take sides during the Trojan War: “Man has only himself to blame if his miseries are worse than they ought to be.” Yeah, the biblical “Man,” Neil Armstrong’s “Man”: The word means, of course, “humanity” or “humans,” and it includes both genders, the male and female brains, the Earth-bound “Martians” and “Venusians.”
We’ve long been in the business of trying to create that which mimics what we are: Thinking, responding, and anticipating beings. Statues are nice, but they lack life and only imitate shape. Disney World’s “animatronic characters” imitate shape and movement, but move only to the extent that they are preprogrammed. Factory robots are very practical inventions; car bumpers are no longer misaligned and engines are built with finer tolerances than they were in the pre-robot assembly line.
But robots are at this time incapable of responding to unexpected stimuli on the order of human response, though very good at responding to stimuli we anticipate for them (Did Pavlov know what was to come?). Characters in computer games do things humans can’t, defying the physical laws of the “real” world; they are also two-dimensional and, for all their abilities, can’t move a feather. Artificial intelligences incarnate in a robot bodies are the product of piece-by-piece additions of sensors. At this time, we have some robots doing what we can do here or there, but not, as we, everywhere with everything at anytime. That’s that “responding and anticipating” that I just noted.
This new soft robot technology is a further step toward our making a being in our own image. In creating intelligent, responsive robots, we are like the mythical sculptor Pygmalion, whose statue Galatea came to life. And, now that I come to think of Pygmalion, maybe our goal outside utilitarian technology is to make companions-to-order, male or female “Stepford wives” that carry none of our human unpredictability. Shades of “sexbots”! (Do we create with the ultimate goal of satisfying our hormonal drives?)
As neurologist Oliver Sacks, author of the popular book The Man Who Mistook His Wife for a Hat: And Other Clinical Tales, noted, humans are a “miracle of integration.” That integration results in a sense of self; otherwise, we are merely more sophisticated robots that respond to our environments as biological computers. In our attempt to recreate ourselves in the form of robots, we fail to consider the principle of integration because it is too difficult to imitate. And there’s something else beyond the principle of integration: The Principle of Gender Difference. As Louann Brizendine explains in The Female Brain, there are in general identifiable differences in “Man” that Zeus, from his perch on Olympus, glossed over in his comment. Female brains, according to Brizendine, contain on average more neurons (11% more) devoted to language and hearing than men, and they have a slightly larger hippocampus. In an on-average relatively smaller female head, female brains also pack more densely the same number of neurons as men with an on-average larger (more voluminous) head—yes, guys, you have bigger heads, but you incorporate more “emptiness.” There’s a difference, even if an on-average difference. And that difference will probably keep us a step ahead of robot we create, even soft ones imbued with some very fast processors.
Creating a “human-like robot” is difficult, but, were we to do so, therein lies the danger that Stephen Hawking and others—many of them science fiction writers—tell us to avoid. When Rollbot can both change shape and employ AI while responding to different stimuli, we’re in at least a little trouble of our losing that place at the head of the hierarchy. Rollbot’s successors will have none of our frailties unless we, the creators, build them into their makeup.
But gender is a problem for robot creators. Toshiba Corp.’s Aiko Chihira, the Japanese-speaking and very female-looking robot that directed customers at the Bank of Tokyo Mitsubishi UFJ and the Mitsukoshi store in Nihonbashi, Tokyo, was a long way from being “human.” That Aiko Chihira looked human was simply a capitulation to our need to think “she” was human. A robot in the shape of a box could give the same information as the robot with human features and multiple human-like expressions. ** Aiko (if I may be so bold as to refer to “her” by her first name) is “humanoid” at best, more appearance than substance, with remarkably life-like features, but without “life.” “She” has no sense of self, and we cannot assume that she “thinks like a woman.”
If you look at Aiko and other robots, you see in them our limitations in making beings in our own image. Of course, the goal of robotics engineers is to make a robot that can sense its environment as we do and respond to stimuli. To do so means they would have to program robots for the unexpected vicissitudes of the real-world environment; and when the engineers fail to anticipate what the robot surrogate will encounter, the robot will fail. But notice that engineers design by stereotype. “Human-like” and “humanoid” images shape their engineering designs. But there are those subtle differences in “Man,” and those differences are embedded in the brains of two genders, making imitating what we all are as humans very difficult to mimic.
Pygmalion’s Galatea both looked and acted like a human female. That ancient Greek myth contains a number of assumptions, not the least of which is that a statue can become “woman-like” in the real world of marriage, birth, and motherhood (Galatea’s child was Paphos). In contrast Watson, the IBM Jeopardy champion (against Ken and Brad in 2011) had no human physical features, but outperformed the two greatest trivia champs.*** Watson had to deal only with language and historically recorded knowledge, and it seemed to be capable of making inferences to reach correct answers. But could Watson “feel”? Did it reach its answers in a linear, single-plane, though rapid, set of logical steps while Ken and Brad acquired their answers as aggregates that just popped into their consciousness through multi-planal thinking?
You might argue that as physicists and engineers approach quantum computing, soft robots will become more human-like. But there’s still going to be a difference. The cold machine computing on a quantum level will always be different from the hot brain computing that operates on as yet not fully understood multiple levels of interconnections with a central command that chooses what to attend to and what to ignore. Think of a computer’s processes as occurring on a single plane and a human’s processes as occurring on multiple planes, many of them intersecting. True, the computer can process faster than the brain, but that doesn’t make it more complex or capable of “seeing” what the brain sees in every situation.
Plus, humans aren’t really good at predicting on the basis of mathematical computation, so they assess differently from computers, which are good at processing probabilities. Give me an intuitive robot, and you’ll give me a more human-like robot. Is our robot-goal, then, to make an error-free human?
So, which “image” of “Man” is the image model for “Man”-like robots? The brain in general and all the physical properties associated with it, just the brain’s processing, or the processing and behaviors associated with and connected to stimuli and the brains of others?
Here’s my own “Turing Test”: Make an angry robot. Or a loving one. The motion picture I Am Mother, written by Michael Lloyd Green and Grant Sputore, puts a “loving,” caring robot in charge of rebooting the human race. Mother is even willing to act as a mother would (I won’t give a spoiler, but will recommend the film). But think about it, to make a human, you really have to make a human. All our efforts to create in our own image will fall short, regardless of what the science fiction writers tell us and the engineers promise to make.
And what about those sheets of polymers that seem to respond to the environment? Well, you, too, have polymers. Teamed up with Stuart Hameroff, Roger Penrose explained in his Shadows of the Mind that consciousness derives from superposed quantum states, each acting in its own spacetime geometry. Somehow, they propose, the human brain operates on a level of quantum gravity in tubular polymer structures that can be found inside neurons. Regardless of the somewhat unprovable nature of that explanation, one has to acknowledge that we don’t operate on purely logical grounds as a machine would. Look inside; there’s much going on, too much to keep track of and anyway, no one has yet been able to keep track of the components of the quantum microworld. That means that the replication or duplication of a thinking, responding, and anticipating being we know as “Man” is beyond our present and future technologies. Why go out of our way to train inorganic polymers to do what those tubular polymer structures in our neurons already do? Again, is it because we have a God Complex?
And so my answer to that question comes back to a theme I’ve hit on before. Remember the Adam and Eve story? If you reduce it to its essence, it’s about pride. They wanted to be like God. And we, in creating robot after robot, refining AI as much as we can, seem to have inherited that ancient human desire. “I want to be like God, and making a thinking, responding, and anticipating being is my path to apotheosis.”
Short of our creating in our own image some robot, we take on the task of shaping others, our children, as best we can to make them in our own image. But, as in creating the perfect human substitute in a robot, creating the perfect replica of us in our children is doomed to failure. In part, that lack of success lies in our thinking of individuals in terms of “Man,” that is, the “general human.” As every parent discovers, there are just too many variables to manipulate, to control, to anticipate. And if Penrose is correct, then we’ll never get a handle on the illogical nature of our quantum nature, where probability, and not predictability, rules.
Of course, the course is set. Engineers will continue to create robot after robot after robot. No one is going to change that unless our technological civilization fails. Ironically, there are those among us those who would, in fact, destroy civilization and their very hope of achieving such godlike status, and many of them—as terrorism of the last twenty years has shown—seek to destroy in the name of their deity. That self-destructiveness is part of “Man’s” potential, and its presence means that to truly create in our own image, we would have to incorporate it in our robot progeny. We have two kinds of brains (in general), unpredictability, and integration of Self that make up individuals and that make a creation of “Man” impossible to achieve no matter how much our desire to be godlike drives us in our creations.
*Caltech. Self-folding “Rollbot” paves the way for fully untethered soft robots. August 21, 2019.
https://www.caltech.edu/about/news/self-folding-rollbot-paves-the-way-for-fully-untethered-soft-robots
**Hongo, Jun. Robotic Customer Service? In This Japanese Store, That’s the Point. The Wall Street Journal online, April 16, 2015. https://blogs.wsj.com/japanrealtime/2015/04/16/toshiba-humanoid-robot-to-debut-in-tokyo-department-store/
*** https://www.youtube.com/watch?v=P18EdAKuC1U
RSS Feed