Some of Gibson’s most fascinating comments were about how our era would be thought about by people in the far future. If the Victorians are known for their denial of the reality of sex, Gibson said, we will be known for our odd fixation with distinguishing real from virtual reality. This comment resonated with me on many different levels. Just a couple weeks before, I had lunch with Craig Mundie, the head of Microsoft Research, prior to a talk he gave at Northwestern. He told us about some new directions they are taking one of their hottest products, the Kinect. The Kinect is a camera for the Xbox gaming system that can see things in 3D. One of their new endeavors with this camera is to allow you to create 3D avatars that move and talk as you are in real time, so you can have very realistic virtual meet-ups. This is now available on the Xbox as Avatar Kinect. The second direction is the real time generation of 3D models of the world around you as you sweep the Kinect around by hand, called Kinect Fusion. With this model of the world around you, you can start to meld real and virtual in some very fun ways. In one of his demos, Mundie waved a Kinect around a clay vase on a nearby table. We instantly got an accurate 3D model up on the screen – exciting and impressive from a $150 gizmo. I’ve had to create 3D models of stuff in my own research, and that’s involved hardware about 100 times more expensive. Even more impressive, Mundie next had the projected image of the 3D model of the vase start to spin, then stuck his hands out in front of the Kinect and used movements of his hand to sculpt it, potter-like. It was wild. All that was needed to complete the trip was a quick 3D print of the result. Further demos showed other ways in which the line between reality and virtuality was being blurred, and it all brought me back to the confluence of real and virtual worlds so well envisioned by the show I advised during its brief life, Caprica.

Gibson’s right. We haven’t yet moved beyond our need to identify what belongs to what when it comes to digital and physical worlds, so we constantly consecrate it with our language. Ironically, some of that very language was created by him: “cyberspace,” a word Gibson coined in his story “Burning Chrome” in 1982. During the conversation today, led by fellow faculty member and author Bill Savage, Gibson said he’s less interested in its rise than to see it die out. He sees its use as a hallmark of our distancing ourselves from who we are as mediated by computer technology. He thinks the term is starting to go out of use, and he’s happy about that — in his view, there’s no need for a word about a space that we are constantly moving through the coordinates of, as we do each time we go on to twitter, facebook, google+, and other digital extensions of self. It’s not cyberspace anymore: it’s our space.

It seemed inevitable that a question about The Singularity would be put to Gibson in the Q&A. Sure enough, it was the final note, and Gibson dispatched it with typical incisiveness. The Singularity, he said, is the Geek Rapture. The world will not change in that way. Like our gradual entrance into cyberspace, now complete enough that marking this world with a separate term seems quaint, Gibson said we will eventually find ourselves sitting on the other side of a whole bunch of cool hardware. But, he feels our belief that it will be a sudden, quasi-religious transformation (perhaps with Cylon guns blazing?) is positively 4th century in its thinking.

The future is impossible to predict. But that’s not going to stop people from trying. We can at least pretend to know where it is we want humanity to go. We hope that laws we craft, the technologies we invent, our social habits and our ways of thinking are small forces that, when combined over time, move our species towards a better existence. The question is, How will we know if we are making progress?

As a movement philosophy, transhumanism and its proponents argue for a future of ageless bodies, transcendent experiences, and extraordinary minds. Not everyone supports every aspect of transhumanism, but you’d be amazed at how neatly current political struggles and technological progress point toward a transhuman future. Transhumanism isn’t just about cybernetics and robot bodies. Social and political progress must accompany the technological and biological advances for transhumanism to become a reality.

But how will we able to tell when the pieces finally do fall into place? I’ve been trying to answer that question ever since Tyler Cowen at Marginal Revolution was asked a while back by his readers: What are the exact conditions for counting “transhumanism” as having been attained? In an attempt to answer, I responded with what I saw as the three key indicators:

1. Medical modifications that permanently alter or replace a function of the human body become prolific.
2. Our social understanding of aging loses the “virtue of necessity” aspect and society begins to treat aging as a disease.
3. Rights discourse would shift from who we include among humans (i.e. should homosexual have marriage rights?) to a system flexible enough to easily bring in sentient non-humans.

As I groped through the intellectual dark for these three points, it became clear that the precise technology and how it worked was unimportant. Instead, we need to figure out how technology may change our lives and our ways of living. Unlike the infamous jetpack, which defined the failed futurama of the 20th century, the 21st needs broader progress markers. Here are seven things to look for in the coming centuries that will let us know if transhumanism is here.

When we think of the future, we think of technology. But too often, we think of really pointless technology – flying cars or self-tying sneakers or ray guns. Those things won’t change the way life happens. Not the way the washing machine or the cell phone changed the way life happens. Those are real inventions. It is in that spirit that I considered indicators of transhumanism. What matters is how a technology changes our definition of a “normal” human. Think of it this way: any one of these indicators has been fulfilled when at least a few of the people you interact with on any given day utilize the technology. With that mindset, I propose the following seven changes as indicators that transhumanism has been attained.

1. Prosthetics are Preferred: The arrival of prosthetics and implants for organs and limbs that are as good as or better than the original. A fairly accurate test for the quality of prosthetics would be voluntary amputations. Those who use prosthetics would compete with or surpass non-amputees in physical performances and athletic competitions. Included in this indicator are cochlear, optic implants, bionic limbs and artificial organs that are within species typical functioning and readily available. A key social indicator will be that terminology around being “disabled”and “handicapped” would become anachronous. If you ever find yourself seriously considering having your birth-given hand lopped off and replaced with a cybernetic one, you can tick off this box on your transhuman checklist.

2. Better Brains: There are three ways we could improve our cognition. In order of likelihood of being used in the near future they are: cognitive enhancing drugs, genetic engineering, or neuro-implants/ prosthetic cyberbrains. When the average person wakes up, brews a pot of coffee and pops an over-the-counter stimulant as or more powerful than modafinil, go ahead and count this condition achieved. Genetic engineering and cyberbrains will be improvements in degree and function, but not in purpose. Any one of these becoming commonplace would indicate that we no longer cling to the bias that going beyond the intelligence dished out by the genetic and environmental lottery is “cheating.”

3. Artificial Assistance: Artificial Intelligence (AI) and Augmented Reality (AR) integrated into personal, everyday behaviors. In the same way Google search and Wikipedia changed the way we research and remember, AI and AR could alter the way we think and interact. Daedalus in Deus Ex and Jarvis in Iron Man are great examples of Turing-quality (indistinguishable from human intelligence) AI that interact with the main character as both side kicks and secondary minds. Think of it this way: you walk into a cocktail party. Your cyberbrain’s AI assist analyzes every face in the room and determines those most socially relevant to you. Using AR projected onto your optic implants, the AI highlights each person in your line of sight and, as you approach, provides a dossier of their main interests and personality type. Now apply this level of information access to anything else. Whether it’s grilling a steak or performing a heart transplant, AI assist with AR overlay will radically improve human functioning. When it is expected that most people will have an AI advisor at their side analyzing the situation and providing instructions through their implants, go ahead and count humanity another step closer to being transhuman.

4. Amazing Average Age: The ultimate objective of health care is that people live the longest, healthiest lives possible. Whether that happens due to nanotechnology or genetic engineering or synthetic organs is irrelevant. What matters is that eventually people will age more slowly, be healthier for a larger portion of their lives, and will be living beyond the age of 120. Our social understanding of aging will lose the “virtue of necessity” aspect and society will treat aging as a disease to be mitigated and managed. When the average expected life span exceeds 120, the conditions for transhuman longevity will have arrived.

5. Responsible Reproduction: Having children will be framed almost exclusively in the light of responsibility. Human reproduction is, at the moment, not generally worthy of the term “procreation.” Procreation implies planned creation and conscientious rearing of a new human life. As it stands, anyone with the necessary biological equipment can accidentally spawn a whelp and, save for extreme physical neglect, is free to all but abandon it to develop in an arbitrary and developmentally damaging fashion. Children – human beings as a whole – deserve better. Responsible reproduction will involve, first and foremost, better birth control for men and women. Abortions will be reserved for the rare accidental pregnancy and/or those that threaten the life of the mother. Those who do choose to reproduce will do so via assisted reproductive technologies (ARTs) ensuring pregnancy is quite deliberate. Furthermore, genetic modification, health screening, and, eventually synthetic wombs will enable the child with the best possibility of a good life to be born. Parental licensing may be part of the process; a liberalization of adoption and surrogate pregnancy laws certainly will be. When global births stabilize at replacement rates, ARTs are the preferred method of conception, and responsible child rearing is more highly valued than biological parenthood, we will be procreating as transhumans.

6. My Body, My Choice: Legalization and regulation will be based on somatic rights. Substances that are ingested – cogno enhancers, recreational drugs, steroids, nanotech – become both one’s right and responsibility. Actions such as abortion, assisted suicide, voluntary amputation, gender reassignment, surrogate pregnancy, body modification, legal unions among adults of any number, and consenting sexual practices would be protected under law. One’s genetic make-up, neurological composition, prosthetic augmentation, and other cybernetic modifications will be limited only by technology and one’s own discretion. Transhumanism cannot happen without a legal structure that allows individuals to control their own bodies. When bodily freedom is as protected and sanctified as free speech, transhumanism will be free to develop.

7. Persons, not People: Rights discourse will shift to personhood instead of common humanity. I have argued we’re already beginning to see a social shift towards this mentality. Using a scaled system based on traits like sentience, empathy, self-awareness, tool use, problem solving, social behaviors, language use, and abstract reasoning, animals (including humans) will be granted rights based on varying degrees of personhood. Personhood based rights will protect against Gattaca scenarios while ensuring the rights of new forms of intelligence, be they alien, artificial, or animal, are protected. When African grey parrots, gorillas, and dolphins have the same rights as a human toddler, a transhuman friendly rights system will be in place.

Individually, each of these conditions are necessary but not sufficient for transhumanism to have been attained. Only as a whole are they sufficient for transhumanism to have been achieved. I make no claims as to how or when any or all of these conditions will be attained. If forced to guess, I would say all seven conditions will be attained over the course of the next two centuries, with conditions (3) and (4) being the furthest from attainment.

Transhumanism is a long way from being attained. However, with these seven conditions in mind, we can at least determine if we are moving towards or away from a transhuman future.

Follow Kyle on his personal blog, Pop Bioethics, and on facebook and twitter.

Image of psychedelic human eye by Kate Whitley via dullhunk on Flickr Creative Commons.

The nerd echo chamber is reverberating this week with the furious debate over Charlie Stross’ doubts about the possibility of an artificial “human-level intelligence” explosion – also known as the Singularity. As currently defined, the Singularity will be an event in the future in which artificial intelligence reaches human level intelligence. At that point, the AI (i.e. AI n) will reflexively begin to improve itself and build AI’s more intelligent than itself (i.e. AI n+1) which will result in an exponential explosion of intelligence towards near deity levels of super-intelligent AI After reading over the debates, I’ve come to a conclusion that both sides miss a critical element of the Singularity discussion: the human beings. Putting people back into the picture allows for a vision of the Singularity that simultaneously addresses several philosophical quandaries. To get there, however, we must first re-trace the steps of the current debate.

I’ve already made my case for why I’m not too concerned, but it’s always fun to see what fantastic fulminations are being exchanged over our future AI overlords. Sparking the flames this time around is Charlie Stross, who knows a thing or two about the Singularity and futuristic speculation. It’s the kind of thing this blog exists to cover: a science fiction author tackling the rational scientific possibility of something about which he has written. Stross argues in a post entitled “Three arguments against the singularity” that “In short: Santa Clause doesn’t exist.”

This is my take on the singularity: we’re not going to see a hard take-off, or a slow take-off, or any kind of AI-mediated exponential outburst. What we’re going to see is increasingly solicitous machines defining our environment — machines that sense and respond to our needs “intelligently”. But it will be the intelligence of the serving hand rather than the commanding brain, and we’re only at risk of disaster if we harbour self-destructive impulses.

We may eventually see mind uploading, but there’ll be a holy war to end holy wars before it becomes widespread: it will literally overturn religions. That would be a singular event, but beyond giving us an opportunity to run [Robert] Nozick’s experience machine thought experiment for real, I’m not sure we’d be able to make effective use of it — our hard-wired biophilia will keep dragging us back to the real world, or to simulations indistinguishable from it.

I am thankful that many of the fine readers of Science Not Fiction are avowed skeptics and raise a wary eyebrow to discussions of the Singularity. Given his stature in the science fiction and speculative science community, Stross’ comments elicited quite an uproar. Those who are believers (and it is a kind of faith, regardless of how much Bayesian analysis one does) in the Rapture of the Nerds have two holy grails which Stross unceremoniously dismissed: the rise of super-intelligent AI and mind uploading. As a result, a few commentators on emerging technologies squared off for another round of speculative slap fights. In one corner, we have Singularitarians Michael Anissimov of the Singularity Institute for Artificial Intelligence and AI researcher Ben Goertzel. In the other, we have the excellent Alex Knapp of Forbes’ Robot Overlords and the brutally rational George Mason University (my alma mater) economist and Oxford Future of Humanity Institute contributor Robin Hanson. I’ll spare you all the back and forth (and all of Goertzel’s infuriating emoticons) and cut to the point being debated. To paraphrase and summarize, the argument is as follows:

1. Stross’ point: Human intelligence has three characteristics: embodiment, self-interest, and evolutionary emergence. AI will not/cannot/should not mirror human intelligence.

2. Singularitarian response: Anissimov and Goertzel argue that human-level general intelligence need not function or arise the way human intelligence has. With sufficient research and devotion to Saint Bayes, super-intelligent friendly AI is probable.

3. Skeptic rebuttal: Hanson argues A) “Intelligence” is a nebulous catch-all like “betterness” that is ill-defined. The ambiguity of the word renders the claims of Singularitarians difficult/impossible to disprove (i.e. special pleading); Knapp argues B) Computers and AI are excellent at specific types of thinking and augmenting human thought (i.e. Kasparov’s Advanced Chess). Even if one grants that AI could reach human or beyond human level, the nature of that intelligence would be neither independent nor self-motivated nor sufficiently well-rounded and, as a result, “bootstrapping” intelligence explosions would not happen as Singularitarian’s foresee.

In essence, the debate is that “human intelligence is like this, AI is like that, never the twain shall meet. But can they parallel one another?” The premise is false, resulting in a useless question. So what we need is a new premise. Here is what I propose instead: the Singularity will be the result of a convergence and connection of human intelligence and artificial intelligence.

Intelligence is extremely hard to define. For the sake of discussion, I’ll define it here as the “ability to analyze a situation, determine a problem, develop a solution, and execute.” As Knapp’s example of Kasparov’s Advanced Chess illustrates, humans and computers are much better than one another at specific elements of chess. A computer is significantly more intelligent when it comes to chess tactics. A human is significantly more intelligent when it comes to strategy. Extrapolation of this analogy (as well as Knapp’s analysis of Watson on Jeopardy!) points towards a human intelligence superiority around abstraction, invention, creativity, and imagination and a computer intelligence superiority in calculation, data analysis, and information retrieval. Thus, I propose a new analogy for the two types of intelligence represented by humans and computers: the right and left hemispheres of the human brain.

It is often said that humans are the animal that can reason. But that description is incomplete. Humans are the animal that can reason creatively and abstractly, or perform the inverse, imagine logically and rationally. To my knowledge (I’d love to be corrected) computers and AI algorithms cannot at this point in time replicate any form of right-brain thinking. But computers are orders of magnitude better at short-term, sharp-focus left-brain thinking. Combine this line of thought with the extended brain hypothesis of Andy Clark and the augmentation-based Singularity survival strategy of David Chalmers, and picture of a cybernetic future begins to emerge. Thus, I argue the Singularity should be re-imagined as a cybernetic process in which the human mind is progressively augmented with better and more complimentary artificial left-brain capacities.

As Advanced Chess demonstrates, a human with a computer is far superior to either a human alone or a computer alone. Consider the analogy of Geordi La Forge and the USS Enterprise computer being comparable with Data. Through the Enterprise, Geordi has access to the same vast processing power Data possesses, but also his own creative and inventive capacities that the Enterprise alone cannot mirror. Data’s most “human” moments are when he expresses these right-brain tendencies and are, in fact, what are referenced when defending Data’s personhood. It is what makes him unique and impossible to replicate with ease.

At our current state of technology, smartphones represent the most advanced and prolific form of cybernetic left-brain augmentation. These hand-held exobrains allow us to perform a multitude of processes and recall or access tremendous amounts of information through visual and auditory interfaces. As neuro-interface technology improves (hat tip Greg Fish) the information on the internet and stored in our external brains will become more expansive and more intimately connected with our nervous systems. The steps toward the Singularity will not be progressive improvement of general AI but of the gradual blending of the biological wetware of the human brain with the artificial hardware of computer technology. The Singularity will be the perfection of the mind-computer interface, such that where the mental processes of the human right-brain ends and the high-powered computer left-brain ends will be indistinguishable both externally by objective observation and internally by the subjective experience of the individual. I call this event the Cybernetic Singularity.

The Cybernetic Singularity differs from the AI Singularity in several ways and, in the process, solves several AI conundrums, both of the technological and philosophical variety.

First and foremost, the ethical “can of worms” of making pure AI is eliminated. So long as the person having his or her mind augmented grants rationally informed and deliberative consent, then no breach of ethics occurs. The concern over experimentally creating, shutting off, or restrictively programming a new form of life is eliminated.

Second, the problem of completely replicating the human mind is eliminated. Cybernetic augmentation will enhance those processes of the brain at which computers excel – memory, data analysis, and computation – without needing to replicate aspects of the brain we are barely beginning to understand, like imagination and creativity.

Third, the theological fears and philosophical qualms around uploading will be mitigated by the slow integration and blending process. Theologians can presume the “seat of the soul” rests in the right hemisphere. Because the process is gradual and the self can reflexively begin to include the augmentations into the mind’s “I” construction, the worries over mind-clones and other philosophical oddities are reduced to interesting thought experiments.

Fourth, the technology is feasible. Memory stimulation, cochlear implants, bionic eyes, and haptic interfaces for prosthetics are rudimentary but empirical and existing forms of neuro-computer interfaces.

Fifth, and most relevant for fans of the apocalypse, no “hard take off” or “AI bootstrapping” will occur. In part, because the blending will be gradual as interfaces and technology incrementally improve there will be no one augmented person who is unstoppably or even significantly more “intelligent” than other augmented individuals. Also in part because there will be a human being at the center of the cyber-brain, still able to make ethical decisions and express self-interest that expands to the universal level of humanity’s self-interest.

The final reason I believe the Cybernetic Singularity is more probable than the AI Singularity is simply that it makes more sense. AI’s designed to do very specific tasks that are labor and data intensive make economic sense and are of obvious value, AI’s designed to mirror things humans are naturally good at seems pointless. Humans have augmented our memory, our ability to calculate, and our ability to process data reliably throughout history. We’ve been slowly augmenting our left-hemisphere since the invention of language.

In sum, The Cybernetic Singularity is the logical extension of a process humans have been pursuing throughout history: the augmentation of our brain’s computational left-hemisphere. By recognizing the relative functions of the hemispheres of the human brain, we are able to see how cybernetic augmentation of the left-hemisphere of the human brain will enable significant increases in some forms of intelligence. Pure general AI is not necessary for an intelligence increase. My theory of The Cybernetic Singularity reconciles the exponential increase in computing technology with the tremendous hurdles facing AI and overcomes the ethical, philosophical, and theological concerns around uploading and the creation of AI and/or mind uploading. The result is a human future that we can reasonably, incrementally, and ethically pursue.

Follow Kyle on his personal blog, Pop Bioethics, and on facebook and twitter.

Image via theWarehouse

Hat tip to Futurismic for many of the links.

Health care is broken. In the US quality of care is tanking. Even in countries with successful universal health care systems costs are rising too fast for the systems to cope. So what do we do?

Atul Gawande, who knows a thing or two about improving healthcare, argues in his commencement address to Harvard that doctors need pit crews:

We are at a cusp point in medical generations. The doctors of former generations lament what medicine has become. If they could start over, the surveys tell us, they wouldn’t choose the profession today. They recall a simpler past without insurance-company hassles, government regulations, malpractice litigation, not to mention nurses and doctors bearing tattoos and talking of wanting “balance” in their lives. These are not the cause of their unease, however. They are symptoms of a deeper condition—which is the reality that medicine’s complexity has exceeded our individual capabilities as doctors.

Gawande has two main arguments. First, that when doctors use checklists they prevent errors and quality of care goes way up. Second, that doctors need to stop acting like autonomous problem solvers and see themselves as a member of a tight-knit team. Gawande is one of the few sane voices in the health care debate. However, later on in his speech, he says that the solution to the health care conundrum is not technology. To a large degree, I agree with him. But not completely. Tech still has a big role to play. If we take a closer look at Dune and Star Trek, we’ll see why Qualcomm and the X-Prize Foundation are ponying up 10 million bucks to fund a piece of medical technology that could help make Gawande’s dream of team-based medicine a bit closer to becoming reality.

In Star Trek: The Next Generation, Beverly Crusher is responsible for a starship with just over a thousand crew members of varying ages and species. Sickbay is, however, not manned by a huge number of staffers. Normally it’s just Dr. Crusher and an assistant or two. Furthermore, Crusher is no Gregory House MD. She lacks both his encyclopedic mind and his caustic personality. Yet Crusher is able to handle a hypothetical complexity that should blow to smithereens anything current doctors could possibly face. How?

The X-Prize Foundation has some ideas – Crusher has a few pieces of tech that let her treat the patient instead of requiring her to be an all-in-one interspecies diagnostician, surgeon, disease knowledge database, and bedside manner superstar. Two tools – the tricorder and the ship’s computer – enable her to access a huge amount of precise data and then compare every known condition or disease against that data to find relevant and probable causes. Qualcomm has teamed up with the X-Prize foundation to fund what they call the Tricorder X-Prize. I wrote about the prize when it was first in the works a while ago. At that time, the prize was called the A.I. physician X-Prize. A new press release renamed the prize and Gawande’s Harvard address has cast the competition ($10 million are at stake) in a new light.

The goal of the prize is for a team “to develop a mobile solution that can diagnose patients better than or equal to a panel of board certified physicians.” Thanks to cloud computing and ubiquitous internet access, pretty much any smartphone can access a server-based data-bank of medical diagnostic information. The trick is to make symptom and data input consistent and accurate, such that the information can be processed and compared against the database. On first glance, it seems the prize may be misnamed. The tricorder only collects data, it is Crusher and the computer that diagnose the disease.

Yet for the solution to be a success, the mobile solution has to be able to, in a sense, force the person utilizing it to become the tricorder. That’s where we come back to Gawande and his justified love affair with checklists. My suspicion is that the winning team will use a checklist based interface to ensure the human-based proxy-tricorder gets all the details and data necessary to ensure a proper diagnosis.

For the Dune fans out there, the Qualcomm X-Prize may sound neither like an A.I. nor a Tricorder, but a digital mentat. Sentient or “thinking machines” are anathema to the inhabitants of the Dune Universe. As a result some human individuals undergo intense conditioning to train their brains in processing data and finding the most logical solution supported by the data provided. These individuals are called mentats. Give enough good data to a mentat and the mentat will provide the right answer. Mentats are used by everyone, particularly governments, to calculate outcomes of complex political decisions. Digital mentat may sound oxymoronic, but recall that a mentat is merely a person trained to achieve the skills of a computer without the risk of sentience. Data processing is the simplest of mentat tasks, but requires a level of mental acumen that is probably impossible in current real humans.

But consider who would be most qualified for this data collection and entry: nurses and physicians. Instead of burdening themselves with a crushing cognitive load of patient history, symptoms, measurements, test-results, and nuances of reporting, the doctor could focus on beside manner, coaching the patient’s treatment and seeking the most accurate and complete information possible. The digital mentat, given sufficient data, could do the diagnosis on its own, just as the ship’s computer does for Beverly Crusher.

Whether it comes in the form of a mobile solution, or even a piece of software connected to a secure database that coordinates symptom and data input, doctors are going to need a medical data processor on their pit crew. The Qualcomm X-Prize “mobile solution” will be a kind of medical mentat, able to deduce the diagnosis from the data provided. To get that data, the doctor and the rest of the medical team will have to become the Tricorder. Whether or not Qualcomm’s Tricorder X-Prize will make that happen, I just don’t know.But the $10 million Ansari X-Prize was enough to jump start a second space race. Maybe a combo of tech and teamwork will be enough to turn health care around.

Follow Kyle on his personal blog and on facebook and twitter.

Image via the X-Prize Foundation Qualcomm X-Prize press release page. Also strikingly familiar to the image I whipped up for my first article on the Tricorder X-Prize.

I am a scientist and academic by day, but by night I’m increasingly called upon to talk about transhumanism and the Singularity. Last year, I was science advisor to Caprica, a show that explored relationships between uploaded digital selves and real selves. Some months ago I participated in a public panel on “Mutants, Androids, and Cyborgs: The science of pop culture films” for Chicago’s NPR affiliate, WBEZ.  This week brings a panel at the Director’s Guild of America in Los Angeles, entitled “The Science of Cyborgs” on interfacing machines to living nervous systems.

The latest panel to be added to my list is a discussion about the first transhumanist opera, Tod Machover’s “Death and the Powers.” The opera is about an inventor and businessman, Simon Powers, who is approaching the end of his life. He decides to create a device (called The System) that he can upload himself into (hmm I wonder who this might be based on?). After Act 2, the entire set, including a host of OperaBots and a musical chandelier (created at the MIT Media Lab), become the physical manifestation of the now incorporeal Simon Powers, who’s singing we still hear but who has disappeared from the stage. Much of the opera is exploring how his relationships with his daughter and mother change post-uploading. His daughter and wife ask whether The System is really him. They wonder if they should follow his pleas to join him, and whether life will still be meaningful without death. The libretto, by the renown Robert Pinsky, renders these questions in beautiful poetry. It will open in Chicago in April.

These experiences have been fascinating. But I can’t help wondering, what’s with all the sudden interest in transhumanism and the singularity?

The media is so saturated with the claim that the Singularity will arrive by 2045 that skeptics are by default on the defensive. Worth noticing amidst the rancor is a recent result by friend and colleague Konrad Kording, who just showed that the number of neurons that we can simultaneously record from is following Moore’s Law. Not long ago, we were limited to recording the activity of a single brain cell at a time; more recently, we can record from several hundred at once. When you examine the trend over 56 different studies, Kording and his student showed that the number is doubling every seven years. Although this is a longer interval than Moore’s Law (two year doublings), what’s really important is that the growth is exponential. Exponential growth lies at the heart of the arguments for the nearness of the Singularity. Given Kording’s result, however, how long do you think it will be before we can record from every neuron in the brain at once? You might be surprised: even with this incredible exponential growth, it will take 220 years. If we suppose that uploading our consciousness will at a minimum entail recording the pattern of activity of the entire brain (why not–it’s no less plausible than every other argument out there), then we can’t even get cracking until 2231.

Of course, the time of the Singularity is not the time when we can upload consciousness, but rather when we create super-intelligent machines (which, according to some, will then devote themselves to figuring out how to beat aging and upload our consciousness, rather than chasing us to the ends of the galaxy). Whether 2045 is reasonable is hotly debated. I expect it’s on the short side by a century or so–but as someone who often thinks in evolutionary time scales, I still view this as an inconsequential amount of time.

But if we weigh the evidence for when the Singularity will occur versus the evidence for world-wide environmental destruction (such as that we’re now exceeding three of ten “planetary boundaries” for sustainable human existence), it’s pretty clear that these threats to our continued existence as a species are looming far faster on the horizon than either the Singularity or uploaded immortality.

So what’s going on? Is environmentalism “tired” and transhumanism “wired”? Is transhumanism just a fleeting new fascination like colonizing space was not long ago, and this soon will also pass? Or is there something more primal going on?

As I pondered these questions recently, it occurred to me that perhaps the transhumanism trend has something to do with secular people–as scientists, engineers, and sci-fi fans tend to be–having an outlet for talking about things that people with religion have more established frameworks for expressing.

Consider this: Scott Atran, among others, has argued that the urge for religion has an evolutionary basis, rooted in our fears of death and predators. Since Darwin, if not before, it’s become increasingly difficult, though, for scientifically-minded people to put stock in religion. Added to this, it’s difficult to have conversations in public about religion, not least because we live in a multi-denominational society where the public expression of creed can be viewed as exclusionary. It’s simply not politically correct in many instances. What if the reason for the rapid spread of Singularity and transhumanism talk is that it’s giving people a secular outlet for thinking through their fears of death and dreams of immortality?

A great deal has been written about relationships between religion and transhumanism. Much of it has drawn parallels between transhumanism and religion. But I don’t think that transhumanism is trying to be a religion: I think that it’s giving secularists (like me) an opportunity to talk publicly about death, the afterlife, and the strange puzzles of personal identity that will someday arise in transforming ourselves into cyborgs, copies of our original selves, or fully digital beings (which I’ve explored here, here, and here). It is letting us safely explore these ideas in a less morose way than the typical meat-to-worms narrative to which secularists are usually limited. In doing so, perhaps it is filling a void that religion used to fill but no longer can for many of us.

Image of cylon by Shawn Sharp, from DVICE’s steampunk cylon contest, via GIZMODO.

Plot from “How advances in neural recording affect data analysis,” by Ian H. Stevenson and Konrad P. Kording, in Nature Neuroscience. Published online 26 January 2011; doi:10.1038/nn.2731.

Was it just me, or was their something faintly bizarre about yesterday’s historical ass whooping of man by machine? Maybe it was Brad Rutter’s increasingly frantic swaying as Watson took his lead and asked for yet another clue in its stilted, strangely mis-timed way. Perhaps it was the effect of the last corporate stiff of the event – in front of a stone wall backdrop that seemed a parody of cheesy corporate décor – telling us where Watson’s winnings will go, all while speaking with a monotone that would make Al Gore jealous. Or maybe it was Alex Trebek’s nonchalance after the historic event as he immediately turned his attention to pitching the next day’s all-teen tournament. Somehow I expected balloons and confetti to descend from the ceiling, maybe with the voice of Hal in the background—“I’m sorry Ken, but you were really improving from your performance yesterday. Would you mind taking out the garbage?” The most important intelligence test of machine versus man in decades sails by with hardly the rattle of a plastic fern.

Besides the very impressive technical achievement of Watson, IBM should be congratulated for managing to turn three episodes of Jeopardy! into a three-episode-long infomercial for their brand. We saw breathless executives tell us how Watson was a real game-changer for medicine, genomics, and spiky hairdos for avatars. We saw the lead engineers puzzling over mathematical squiggles written on staggered layers of sliding glass panels (something we’ve seen in an Intel commercial before when it was necessary for a visual joke to work, and so obviously useless for doing real work that it seems an insult to viewers in this context).

The overall feel of the event was highly corporate. Alex Trebek channeled Mikael Blomquist’s obsessiveness over computer model names as he explained how Watson’s brain was a massive cluster composed of several cabinets of IBM Power 750 Servers. I wondered how many takes it took for him to get the spiel down. Amidst the heavily rehearsed corporate messaging, we did get some nuggets of interesting information, like how Watson was initially dumb to gender before, as one of the researchers put it, “it got the gender module.” I’m fairly confident this came in the form of a small cheesecloth bag of genetically modified goat genitalia inserted into the head node of the aforementioned Power Server cluster.

February 16, 2011, will go down in history as the date of a very important milestone in artificial intelligence. As I blogged about earlier, in reaching a machine with the kind of intelligence we want, having goal posts that are at points short of that is extremely helpful, and The Jeopardy Test seems to fit the bill. Beating the human Jeopardy-savants on Wednesday was at turns dramatic and eerie. I think IBM has a major achievement on its hands. I just wish the whole thing had been done with a bit more of a sense of humor, and a bit less gratuitous corporate messaging.

People who work in robotics prefer not to highlight a reality of our work: robots are not very reliable. They break, all the time. This applies to all research robots, which typically flake out just as you’re giving an important demo to a funding agency or someone you’re trying to impress. My fish robot is back in the shop, again, after a few of its very rigid and very thin fin rays broke. Industrial robots, such as those you see on car assembly lines, can only do better by operating in extremely predictable, structured environments, doing the same thing over and over again. Home robots? If you buy a Roomba, be prepared to adjust your floor plan so that it doesn’t get stuck.

What’s going on? The world is constantly throwing curveballs at robots that weren’t anticipated by the designers. In a novel approach to this problem, Josh Bongard has recently shown how we can use the principles of evolution to make a robot’s “nervous system”—I’ll call it the robot’s controller—robust against many kinds of change. This study was done using large amounts of computer simulation time (it would have taken 50–100 years on a single computer), running a program that can simulate the effects of real-world physics on robots.

What he showed is that if we force a robot’s controller to work across widely varying robot body shapes, the robot can learn faster, and be more resistant to knocks that might leave your home robot a smoking pile of motors and silicon. It’s a remarkable result, one that offers a compelling illustration of why intelligence, in the broad sense of adaptively coping with the world, is about more than just what’s above your shoulders. How did the study show it?

Each (simulated) robot starts with a very basic body plan (like a snake), a controller (consisting of a neural network that is randomly connected with random strengths), and a sensor for light. Additional sensors report the position of body segments, the orientation of the body, and ground contact sensors for limbs, if the body plan has them. The task is to bring the body over to the light source, 20 meters away.

A bunch of these robots are simulated, and those that do poorly are eliminated, a kind of in-computo natural selection. The eliminated robots are replaced with versions of the ones that succeeded, after random tweaks (“mutations”) to these better controllers have been made. The process repeats until a robot that can get to the light is found. So far, there’s been no change in the shape of the body.

With the first successful robot-controller combination found (one that gets to the light), the body form changes from snake-like to something like a salamander, with short legs sticking out of the body. (All body shape changes are pre-programmed, rather than evolved.) The evolutionary process to find a successful controller-bot combination repeats, with random changes to the better controllers until, once again, a controller-bot combination is found that is able to claw its way to the light.

Then the short legs sticking out to the side slowly get longer, and rather than sticking out to the side, they progressively become more vertical. With each change in body shape, the evolutionary process to find a controller repeats. Eventually, the sim-bot evolves to something that looks like any four-legged animal.

That was all for round one of evolution. For round two, the best controller from round one was copied into the same starting snake-like body type that round one began with. But now, the change in body forms occurs more rapidly, so that by the time 2/3 of the “lifetime” of the robot is completed, it has reached its final dog-like form. For round three, this all happens within 1/3 of the robot’s lifetime. For round four, the body form starts off as dog-like and stays there.

So there are changes occurring at two different time scales: changes over the “lifetime” of the robot, similar to our own shape changes from fetus to adulthood; and changes that occur over generations, through which development during a lifetime occurs more rapidly. The short time scale is called “ontogenetic” and the long scale (between the different rounds) is “phylogenetic.”

The breakthrough of the work is that it found that having these variations in body shape occur over ontogenetic and phylogenetic time scales resulted in finding a controller that got the body over to the light much faster than if no such changes in body shape occurred. For example, when the system began with the final body type, the dog-like shape, it took much longer to evolve a solution than when the body shapes progressed from snake-like to salamander to dog-like. Not only was a controller evolved more rapidly, but the final solution was much more robust to being pushed and nudged.

The complexity of the interactions over 100 CPU years of simulated evolution makes the final evolved result difficult to untangle. Nonetheless, there is good evidence that the cause of accelerated learning in the shape-changing robots is that the controllers developed through changing bodies have gone through a set of “training-wheel” body shapes: a robot starting with a four-legged body plan and a simple controller quickly fails—it can’t control the legs well and simply tips over. Starting with something on the ground that slithers, as was the case in these simulations, is less prone to such failures. So not any old sequence of shape changes works: mimicking the sequence seen in evolution garners some of the advantages that presumably made this sequence actually happen in nature, such as higher mechanical stability of more ancient forms.

Less clear is the source of increased robustness—the ability to recover from being nudged and pushed in random ways. Bongard suggests that the increased robustness of controllers that have evolved with changing body shapes is due to those controllers having had to work under a wider range of sensor-motor relationships than the ones that evolved with no change in body shape. For example, any controller that’s particularly sensitive to a certain relationship between, say, a sensor that reports foot position, and one that reports spine position would fail (and thus be eliminated) as those relationships are systematically changed in shifting from salamander-like to dog-like body form and movement. So that means that if I suddenly pushed down the back of a four-legged dog-like robot, so that its legs would splay out and it would be forced to move more like a salamander, the winners of the evolutionary competition would still be able to work because the controllers had worked in salamander-like bodies as well as in dog-like bodies.

In support of this idea, the early controllers, that were purely based on moving the body axis (“spine”), appear to be still embedded in the more advanced controllers; so if something happens to the body (say, one leg gets knocked), the robot can revert to more basic spine-based motion patterns that don’t require precise limb control. Bongard observed that the controllers evolved through changing body shape exhibited more dependence on spinal movement, using the legs more for balance, than those evolved without changing body shape. (It would be interesting to try his approach with simulated aquatic robots, which can be neutrally buoyant like many aquatic animals are, and thus don’t have the “tipping over” problem that Bongard’s simulated terrestrial robots had).

To be fair to existing robots, even with a controller that worked under every conceivable body shape and environmental condition, they would still break all the time. This is because the materials we make them out of are not self-healing, in contrast to the biomaterials of animals. Animals are also constantly breaking (at least on a micro level), and the body constantly repairs this. Bones subjected to higher loads, like the racket arm of a tennis player, get measurably thicker. Not only is the body self-repairing, recent innovative computer simulations of real neurons that generate basic rhythms like walking and chewing have shown that the neurons keep generating the rhythm despite big variations in the functioning and connections of these neurons. These functions are so important to continued existence—the body’s version of too big to fail—that embedded within them are solutions to just about everything the world can throw at them.

This new work provides the fascinating and useful result that fashioning controllers that work through a sequence of body shapes mimicking those seen in evolution accelerates the learning of new movement tasks and increases robustness to all the hard knocks that life inevitably delivers. It suggests that without the sequence of body shapes that evolution and development bring about, we might have nervous systems that are much too finely tuned to our adult upright bipedal form. Instead of crawling to help after we twist our ankle in the woods, we’d be left with nothing but howling for help.

During the competition, each of four judges will type a conversation with one of us for five minutes, then the other, and then will have 10 minutes to reflect and decide which one is the human. Judges will also rank all the contestants—this is used in part as a tiebreaking measure. The computer program receiving the most votes and highest ranking from the judges (regardless of whether it passes the Turing Test by fooling 30 percent of them) is awarded the title of the Most Human Computer.

What makes the competition so intriguing is that, as all contestants are ranked, be they human or computer, there is not only an award for the Most Human Computer, but also an award for the Most Human Human. Brian Christian is one of the vetted few humans who has earned the accolade. He describes his experience in the competition in his outstanding article “Mind vs. Machine” in The Atlantic. The article presents a snippet of what will surely be a wonderful book, The Most Human Human.

Like Sherry Turkle, Christian argues that machines are calling our humanity into stark relief. Yet he sees human-like computers not as automatons dragging us into banality, but as imperfect mirrors, reminding us of what makes us human by what they cannot reflect. I suspect it’s Christian’s double-life as a science journalist and poet that drew him to consider our dual-natured human brain:

Perhaps the fetishization of analytical thinking, and the concomitant denigration of the creatural—that is, animal—and bodily aspects of life are two things we’d do well to leave behind. Perhaps at last, in the beginnings of an age of AI, we are starting to center ourselves again, after generations of living slightly to one side—the logical, left-hemisphere side. Add to this that humans’ contempt for “soulless” animals, our unwillingness to think of ourselves as descended from our fellow “beasts,” is now challenged on all fronts: growing secularism and empiricism, growing appreciation for the cognitive and behavioral abilities of organisms other than ourselves, and, not coincidentally, the entrance onto the scene of an entity with considerably less soul than we sense in a common chimpanzee or bonobo—in this way AI may even turn out to be a boon for animal rights.

Among many conclusions Christian draws is that to be more human you must be yourself. But this is no idle command. The process of being oneself is an active, conscious, and, in some cases, laborious task. Consider your average conversation at a cocktail party – safe topics, non-confrontational questions, scripted answers. Part of Christian’s message, it seems, is not that we should worry about a computer sounding human, but that we humans may make the task too easy. So go forth and be quirky, odd, unique, expressive, honest, clever, eccentric, and above all yourself; in a phrase, be more human.

Image of The Most Human Human via RandomHouse

Can you have an emotional connection with a robot? Sherry Turkle, Director of the MIT Initiative on Technology and Self, believes you certainly could. Whether or not you should is the question. People, especially children, project personalities and emotions on to rudimentary robots. As the Chronicle of Higher Education article on her shows, the result of believing a robot can feel is not always happy:

One day during Turkle’s study at MIT, Kismet malfunctioned. A 12-year-old subject named Estelle became convinced that the robot had clammed up because it didn’t like her, and she became sullen and withdrew to load up on snacks provided by the researchers. The research team held an emergency meeting to discuss “the ethics of exposing a child to a sociable robot whose technical limitations make it seem uninterested in the child,” as Turkle describes in [her new book] Alone Together.

We want to believe our robots love us. Movies like Wall-E, The Iron Giant, Short Circuit and A.I. are all based on the simple idea that robots can develop deep emotional connections with humans. For fans of the Half-Life video game series, Dog, a large scrapheap monstrosity with a penchant for dismembering hostile aliens, is one of the most lovable and loyal characters in the game. Science fiction is packed with robots that endear themselves to us, such as Data from Star Trek, the replicants in Blade Runner, and Legion from Mass Effect. Heck, even R2-D2 and C-3PO seem endeared to one another. And Futurama has a warning for all of us.

Yet these lovable mechanoids are not what Turkle is critiquing. Turkle is no Luddite, and does not strike me as a speciesist. What Turkle is critiquing is contentless performed emotion. Robots like Kisemet and Cog are representative of a group of robots where the brains are second to bonding. Humans have evolved to react to subtle emotional cues that allow us to recognize other minds, other persons. Kisemet and Cog have rather rudimentary A.I., but very advanced mimicking and response abilities. The result is they seem to understand us. Part of what makes HAL-9000 terrifying is that we cannot see it emote. HAL simply processes and acts.

On the one hand, we have empty emotional aping; on the other, faceless super-computers. What are we to do? Are we trapped between the options of the mindless bot with the simulated smile or the sterile super-mind calculating the cost of lives?Turkle’s primary concern, and I believe it is a legitimate one, is communication and technology’s influence upon it. Return to one of my favorite pieces of technology, the smartphone (or app phone, as David Pogue describes it). How often have you found yourself uncontrollably pulling the damn thing out of your pocket to check it, regardless of the situation. I have been at parties with friends, at intimate dinners, at funerals, in meetings, and even presenting for class and had to fight the urge to see which piece of junk email set off vibrations in my pocket. Every spare moment it seems like I’m checking Facebook or Instagram or Twitter or email to see what decontextualized scrap of communication I can consume to slake my thirst for that simple thing communication is supposed to create – community.

And therein we find the crux of the matter. Our communication through a lot of technology is not communication at all. It’s one-way “shouting into the void.” Like the defecting Red October we send out one sonar ping, hoping for one ping, just one ping, in return. Kismet and Cog provide that ping, that perfect response, immediately and wordlessly. Every day it seems like there is an article on how to read body language, to understand the wordless communication that happens every day in every interaction. Turkle’s reaction to Kismet and Cog dovetails nicely with Haraway’s lovely line, “Our machines are disturbingly lively, and we ourselves are frighteningly inert.”

Kismet and Cog are designed to mimic genuine body language. As such, they represent a kind of complementary Turing test. The original test was to see if a human could distinguish between conversation with a human and an A.I. by chatting over a computer terminal. The Turkle test, as we might call it, would be the ability to form a genuine relationship with a human – to show concern, to mirror emotion, to recognize the other self. The Turing test tells us if an A.I. can think like a human, the Turkle test tells us if an A.I. can communicate like a human.

The first group of robots to be subjected to the Turkle test may be just around the corner:

“There are advantages to it not being a person—robots can be seen as not judgmental; people are not at risk of losing face to a robot,” [leader of Kismet project, Cynthia] Breazeal says. “People may be more honest and willing to disclose information to a robot that they might not want to tell their doctor for fear of sounding like a ‘bad’ patient. So robots working with other people can help the patient and the care staff.”

During her research, Turkle visited several nursing homes where resi dents had been given robot dolls, including Paro, a seal-shaped stuffed animal programmed to purr and move when it is held or talked to. In many cases, the seniors bonded with the dolls and privately shared their life stories with them.

“There are at least two ways of reading these case studies,” she writes. “You can see seniors chatting with robots, telling their stories, and feel positive. Or you can see people speaking to chimeras, showering affection into thin air, and feel that something is amiss.”

Some robotics enthusiasts argue that these sociable machines will soon mature, and that new models may one day be judged as better than humans for many tasks. After all, robots don’t suffer emotional breakdowns, oversleep, or commit crimes.

Hospice and end-of-life care requires a daily heroic effort. Though I do not deny there are some fantastic and wonderful caretakers, but it would require superhuman abilities to provide a listening, understanding ear to each and every person one is helping live day-to-day. Hospice and elderly care robots designed to pass the Turkle test might provide a solution. Yes, these robots would not really be listening or understanding in the same way a real human hospice worker would. Further, we would need to make sure we avoided the confusion of the little girl from the beginning, ensuring an understanding that yes, this was a robot and, yes, it might break down. But, if we did that, couldn’t hospice bots provide a huge service? Nursing home animals don’t understand a thing either, yet they provide a measurable medical benefit to patients, as well as a general improvement in mood and quality of life. Why shouldn’t we make robots that do that as well?

I end with a story. My grandfather talks to his cat, Mickey, constantly. Mickey is not the brightest of cats and has pretty much one emotion, which is a mix of hunger and disdain. My grandfather is quite aware that Mickey is neither aware of nor concerned with the fact that there is no more beer in the fridge or that the internet is infuriating. But my grandfather loves telling these thoughts to his cantankerous feline companion. He simply likes to express his thoughts vocally; Mickey is his living breathing diary – no response is expected, no worry needed for Mickey’s opinion or mounting frustration with the tedium of petty complaints and pedantic observations.

My family and I visit my grandfather and grandmother regularly. We love chatting it up, contributing our opinions, arguing, and generally communicating with one another in vigorous discussion. My grandfather always seems to get the first and last word, as is his wont. He is not lacking for human companionship by any stretch of the imagination.

But, Mickey is different. Mickey makes my grandfather happy because he listens. He is an un-judging ear and a comically loyal companion. Mickey would pass the Turkle test. Were I to pick the cat up, pop the hatch on his side, and reveal that Mickey is, in fact, a robo-cat, I don’t think my grandfather would give a damn. And that says something vital.

Rochellys Diaz Heijtza and Sven Pettersson and colleagues raised mice in a germ free environment and compared them to mice raised with normal gut bugs. The researchers found that compared to the germ-free mice, the normal mice had reduced expression of two brain molecules, synaptophysin, and PSD-95, in a region of the brain called the striatum. Correlating with this, the germ-free mice had higher levels of activity, and less anxiety than the mice with the normal complement of gut microbes. Amazingly, they also found that there was a “sensitive period” of exposure — a time before which exposure to the gut bugs mattered, and after which exposure didn’t change the brain any more. This is characteristic of many brain regions such as visual cortex, which needs normal visual input to develop properly and provide normal visual ability. If you provide that normal input after the sensitive period, the brain doesn’t fix itself. The scientists found that exposing the germ-free mice to normal gut microbes up to about 6 weeks of age resulted in normal levels of movement and anxiety; but exposure after that age resulted in no change.

How can this be? The paper has some specific technical suggestions, but if you think broadly about animals versus plants, it isn’t completely surprising. Next time you are eating your salad, consider how it is that you ended up eating your greens, rather than the greens eating you. It’s a story that’s almost two billion years in the making.

About 1.6 billion years ago, animals and plants went on their separate ways. One type of organism has the “stay in place and absorb” energy strategy. These are the plants, which sit and photosynthesize all day. The other organism has the “go around and get it” energy strategy – that’s you. The innovation of being an animal, in comparison to plants, is to have a gut with an ability to move, and a nervous system to detect the next good thing to put into that gut and then control the movement system to get the gut to the food.

The correspondence between a mobile gut and having a nervous system is so deep that some animals that give up mobility later in life also lose their nervous system. Ironically, they digest it. This is the tunicate, an animal that swims around in early life, but once they mature, they find a place to settle down on the ocean floor. Having done that, they digest most of their nervous system (some have compared this to getting tenure).

So, it is not a big surprise that key neurotransmitters like serotonin (most of which is excreted by cells in the gut wall in response to food), dopamine, glutamate, GABA, and norepinephrine are heavily represented in the gut, or that the gut is equipped with its own nervous system that has some one hundred million neurons, and almost the same number of types of neurons as the brain (Heribert Watzke has a stimulating TED talk on this).

I know what you’re thinking. This is just a story of how brains came to be. For the purposes of intelligence, the energy may as well come from a portable fusion reactor for all it matters. So any suggestion that AI needs a gut to reach the next level is misguided. I’ll argue that this viewpoint is overly simplistic.

Years ago the philosopher Patricia Churchland and the computational neuroscientist Terrence Sejnowski wrote a book called “The Computational Brain.” In it, they made a striking point regarding a pervasive belief in the AI community regarding the study of the brain. Most of the AI community view the key cognitive powers they are chasing as logically independent of any specific implementation. That is, it’s a formal system they are trying to uncover, and whether that formal system is encoded in silicon, punch cards, a hydraulic machine, biological material, or whatever, does not matter, just as whether the pieces of a chess game are made out of plastic or wood, or pictures on a computer screen, doesn’t matter. Because of this—the multiple realizability of formal systems–some people in AI believe that study the brain is irrelevant.

The brilliant point that Churchland and Sejnowski made was that, although it is true that once you understand the mechanism of the brain, at least certain parts of it may be formally independent of any particular instantiation, the key question for humanity right now is how do we get to this understanding. To get there, they said, we might take our cue from the only existing examples of things that are truly intelligent: animals. We need to study how real examples of intelligence work, crack their mechanism in their full wetware glory, and after that, we can potentially formalize and instantiate in silicon or whatever material we want.

Until recently, most of neuroscience had little inclination to mine questions of how appetitive drives such as hunger, and motivations in general, feed into the rich biomechanical and neuronal story that is being uncovered through the mechanistic study of animals. And yet, as I wrote above, the acquisition of energy through moving the gut around is foundational to “animal-hood” in the first place. Studies like the one showing ties between brain development and the gut testify to the deep interconnections of nervous systems and the guts they evolved to satisfy.

We have every reason to think that a full understanding of gut-brain interactions, and associated reward systems, will lead to a better understanding of how to build an intelligent machine. From this understanding we are also more likely to be able to build machines with the “right” connection between motivations and action, a central issue for people concerned about the consequences of The Singularity for the future of our species.

Image of Olaf Breuning’s “Big Brain Small Stomach” from Arrested Motion

At night in the rivers of the Amazon Basin there buzzes an entire electric civilization of fish that “see” and communicate by discharging weak electric fields. These odd characters, swimming batteries which go by the name of “weakly electric fish,” have been the focus of research in my lab and those of many others for quite a while now, because they are a model system for understanding how the brain works. (While their brains are a bit different, we can learn a great deal about ours from them, just as we’ve learned much of what we know about genetics from fruit flies.) There are now well over 3,000 scientific papers on how the brains of these fish work.

Recently, my collaborators and I built a robotic version of these animals, focusing on one in particular: the black ghost knifefish. (The name is apparently derived from a native South American belief that the souls of ancestors inhabit these fish.  For the sake of my karmic health, I’m hoping that this is apocryphal.) My university, Northwestern, did a press release with a video about our “GhostBot” last week, and I’ve been astonished at its popularity (nearly 30,000 views as I write this, thanks to coverage by places like io9, Fast Company, PC World, and msnbc). Given this unexpected interest, I thought I’d post a bit of the story behind the ghost.

Our first desire for this robot was to provide a kind of telescope. Let me explain. Watching how animals behave so that we can learn how the brain works is challenging. They almost never do the same thing twice and you can’t ask them to please repeat what they just did. But, we scientists love repeatability: only through repetition can we assess whether something is statistically significant. Without repetition, we are at a loss. Enter the robots. With a robot, commanded to move in the same ways as our animal, we can get at previously very difficulty issues. For example, with our robot, we can tell it to repeat a strange and unexpected movement we’ve observed in the fish over and over until we have precisely figured out the underlying mechanical principles of the movement.

So using this approach, we were able to convincingly demonstrate the basis of something quite special. As you know, fish usually swim forward, mostly in a horizontal plane. Our fish is very different. Because it hunts in total darkness and can sense in all directions, it doesn’t really care which way it swims. It’s as agile and responsive swimming backward as it is swimming forward. See below for a sense of how it moves (this high speed video is from my collaboration with George Lauder at Harvard, who shot it):

The ability to swim in the dark forward and backward with equal agility depends on two special abilities: first, the weak electric field I mentioned provides a kind of underwater sonar (think of these fish as underwater bats). Second, the strange way they swim means they can shift from forward to reverse almost instantly. If you look at the photo or video above, you see a long fin along the body. The fish simply wiggles this fin in one direction or the other to swim forward or reverse — all while keeping the body straight! (Very handy if you want to build an underwater robot that is practical, since flexing a body is troublesome to implement.)

Now we watch this fish move forward and backward all the time and we’ve published studies on how it works. But when you watch them, it’s clear they have more tricks than that up their fins. They are quite acrobatic. It really hit us one day when my then student, Oscar Curet, observed the fish moving vertically without any difficulty. But wait! How can a fin, that usually only moves a body forward or back, move a body vertically? This is the mystery that the robot helped us solve—the fish doesn’t do it frequently enough, for a long enough time, for us to really nail it by observation of the animal itself.

So the GhostBot was born. It’s very advanced, with 32 independently controllable motors in a package the size of your forearm (for comparison, industrial robot arms typically have less than 10 independently controllable motors). Because we needed to pack things so tightly, Kinea, a company we are affiliated with who built the robot, designed 32 custom “fish steaks”—circuit boards for each of the 32 motors, which stack together through a spine-like bus connector. These printed circuit board steaks are an integral part of the structure of GhostBot. Here’s what they look like, with one of the motors and a quarter for scale:

A stack of these fish steaks, along with 32 small Swiss-made motors, goes into a cylindrical water proof hull, and then we attach 32 very fine rods and a lycra fin to the bottom to make the artificial ribbon fin. We currently send control signals via a tether so we can move the fin in exactly the way we tell it, allowing us to recreate the fish fin’s motion on command. Finally we get to wash, rinse, and repeat, all while measuring complex mechanical effects in the comfort of our lab.

Here’s a video of the robot swimming in a flow tunnel (an artificial stream). It doesn’t look like it’s moving, because we’ve adjusted the stream speed to match the swimming speed. The rods going up suspend the robot on a frictionless air rail (think of it being suspended from two hovercraft floating on a surface that is out of view), so it is free to move forward, backward, and sideways:

With that done, we can go back and program our robot to do the strange and complex motion we witnessed. The fish sends a wiggle from tail to head, and, at the very same time, a wiggle from head to tail. These two wiggles collide (thankfully, not creating anti-wiggles!) in the middle of the fin. So the propulsion in the horizontal direction is completely canceled out. But the fluid jets that those wiggles had created live on – they collide as well and the result is a spout of fluid going in a downward direction. Like every other animal, this one moves thanks to Newton’s third (to every action, there is an equal and opposite reaction)—so the spout of fluid downward pushes the fish upward.

We can see that spout by putting a bunch of reflective particles in the water, and shining a very powerful laser sheet at it. Here’s what we first saw when we did this, a little over a year ago in George Lauder’s lab (note that the robot is being held rigidly, so we can measure how hard it is pushing up):

A beautiful mushroom-cloud like structure, with an inverted jet. The ghost knife has tamed some very complex fluid mechanics to make the fluid do its bidding. For an animal that hunts in the dark, frequently in large tree root masses alone rivers, the maneuverability this buys it is likely to be essential to its survival.

The GhostBot, and its future offspring, have a promising future. It has a number of compelling attributes compared to other fish for use in robotics, such as being able to near instantaneously change directions, and swimming while keeping the body rigid, making robotic implementation and deployment more practical. Another key advantage arises due to how smoothly the force the fin generates varies as we change things like amplitude of the wiggle down the fin. This adds to other advantages that the robot shares with other fish-based propulsion systems, such as resistance to getting stuck in weeds and other debris, which conventional propulsive technologies like propellers are prone to.

We’ve also built an artificial version of how the fish’s electric sonar works and put that on the robot. We are about to start experiments in which the robot is able to autonomously approach an object, sense it with its electrical field, and then position itself nearby. We are designing a new version of the GhostBot that will have additional capabilities, including fins at the front of the body for pitching and rolling.

Having underwater vehicles powered by such bio-inspired technology will enable new capabilities, such as detailed inspection work in cluttered quarters (a sunken ship, or an exploded well head undersea), and situations where only divers can be used because of the need for up close work near delicate structures (coral reef health monitoring, for example.)

Our ultimate goal is to simultaneously improve our understanding of how to process sensory information for agile movement and develop two stand-alone technologies: robotic undulators for propulsion of highly maneuverable underwater vehicles, and artificial electrosense for use in all kinds of devices where vision may not work. Stay tuned!

For all the details, you can check out the study, published by the Journal of the Royal Society Interface.

I have a confession. I used to be all about the Singularity. I thought it was inevitable. I thought for certain that some sort of Terminator/HAL9000 scenario would happen when ECHELON achieved sentience. I was sure The Second Renaissance from the Animatrix was a fairly accurate depiction of how things would go down. We’d make smart robots, we’d treat them poorly, they’d rebel and slaughter humanity. Now I’m not so sure. I have big, gloomy doubts about the Singularity.

Michael Anissimov tries to restock the flames of fear over at Accelerating Future with his post “Yes, The Singularity is the Single Biggest Threat to Humanity.”

Combine the non-obvious complexity of common sense morality with great power and you have an immense problem. Advanced AIs will be able to copy themselves onto any available computers, stay awake 24/7, improve their own designs, develop automated and parallelized experimental cycles that far exceed the capabilities of human scientists, and develop self-replicating technologies such as artificially photosynthetic flowers, molecular nanotechnology, modular robotics, machines that draw carbon from the air to build carbon robots, and the like. It’s hard to imagine what an advanced AGI would think of, because the first really advanced AGI will be superintelligent, and be able to imagine things that we can’t. It seems so hard for humans to accept that we may not be the theoretically most intelligent beings in the multiverse, but yes, there’s a lot of evidence that we aren’t.

….

Humans overestimate our robustness. Conditions have to be just right for us to keep living. If AGIs decided to remove the atmosphere or otherwise alter it to pursue their goals, we would be toast. If temperatures on the surface changed by more than a few dozen degrees up or down, we would be toast. If natural life had to compete with AI-crafted cybernetic organisms, it could destroy the biosphere on which we depend. There are millions of ways in which powerful AGIs with superior technology could accidentally make our lives miserable, simply by not taking our preferences into account. Our preferences are not a magical mist that can persuade any type of mind to give us basic respect. They are just our preferences, and we happen to be programmed to take each other’s preferences deeply into account, in ways we are just beginning to understand. If we assume that AGI will inherently contain all this moral complexity without anyone doing the hard work of programming it in, we will be unpleasantly surprised when these AGIs become more intelligent and powerful than ourselves.

Oh my stars, that does sound threatening. But again, that weird, nagging doubt lingers in the back of my mind. For a while, I couldn’t place my finger on the problem, until I re-read Anissimov’s post and realized that my disbelief flared up every time I read something about AGI doing something. AGI will remove the atmosphere. Really? How? The article, in fact, all arguments about the danger of the Singularity necessarily presume one single fact: That AGI will be able to interact with the world beyond computers. I submit that, in practical terms, they will not.

Consider the example of Skynet. Two very irrational decisions had to be made to allow Skynet to initiate Judgment Day. First, the A.I. that runs Skynet was debuted on the military network. In the mythos of the film, Skynet does not graduate from orchestrating minor battle plans or strategizing invasions in the abstract, but goes straight from the coder’s hands to getting access to the nuclear birds. Second, in the same moment, the military rolls out a fleet of robot warriors that are linked to Skynet, effectively giving the A.I. hands and then putting guns in those hands.

My point is this: if Skynet had been debuted on a closed computer network, it would have been trapped within that network. Even if it escaped and “infected” every other system (which is dubious, for reasons of necessary computing power on a first iteration super AGI), the A.I. would still not have any access to physical reality. Singularity arguments rely upon the presumption that technology can work without humans. It can’t. If A.I. decided to obliterate humanity by launching all the nukes, it’d also annihilate the infrastructure that powers it. Me thinks self-preservation should be a basic feature of any real AGI.

In short: any super AGI that comes along is going to need some helping hands out in the world to do its dirty work.

B-b-but, the Singulitarians argue, “an AI could fool a person into releasing it because the AI is very smart and therefore tricksy.” This argument is preposterous. Philosophers constantly argue as if every hypothetical person is either a dullard or a hyper-self-aware. The argument that AI will trick people is an example of the former. Seriously, the argument is that  very smart scientists will be conned by an AGI they helped to program. And so what if they do? Is the argument that a few people are going to be hypnotized into opening up a giant factory run only by the A.I., where every process in the vertical and the horizontal (as in economic infrastructure, not The Outer Limits) can be run without human assistance? Is that how this is going to work? I highly doubt it. Even the most brilliant AGI is not going to be able to restructure our economy overnight.

So keep your hats on folks, don’t start fretting about evil AGI until we live in an economy that is solely robot labor. Until then, I just can’t see it. I can’t see how AGI gets hands. Maybe that’s a limit on my vision. But if the nightmare scenario of AGI going sentient and rogue over night comes true, then I think we’re all in good shape. Sure, it might screw up our communications networks, but it’s not going to be able to do much of anything outside a computer. Anytime you start getting nervous, remember all the things we still need people to do, and how much occurs beyond the realm of the computer. In that light, the Singularity is just a digital tempest in a teacup.

Image of a very scary computer bank by k0a1a.net’s photostream via Flickr Creative Commons

Follow Kyle Munkittrick on Twitter @PopBioethics

The Singularity seems to be getting less and less near. One of the big goals of Singularity hopefuls is to be able to put a human mind onto (into? not sure on the proper preposition here) a non-biological substrate. Most of the debates have revolved around computer analogies. The brain is hardware, the mind is software. Therefore, to run the mind on different hardware, it just has to be “ported” or “emulated” the way a computer program might be. Timothy B. Lee (not the internet inventing one) counters Robin Hanson’s claim that we will be able to upload a human mind onto a computer within the next couple decades by dissecting the computer=mind analogy:

You can’t emulate a natural system because natural systems don’t have designers, and therefore weren’t built to conform to any particular mathematical model. Modeling natural systems is much more difficult—indeed, so difficult that we use a different word, “simulation” to describe the process. Creating a simulation of a natural system inherently means means making judgment calls about which aspects of a physical system are the most important. And because there’s no underlying blueprint, these guesses are never perfect: it will always be necessary to leave out some details that affect the behavior of the overall system, which means that simulations are never more than approximately right. Weather simulations, for example, are never going to be able to predict precisely where each raindrop will fall, they only predict general large-scale trends, and only for a limited period of time. This is different than an emulator, which (if implemented well) can be expected to behave exactly like the system it is emulating, for as long as you care to run it.

In short: we know how software is written, we can see the code and rules that govern the system–not true for the mind, so we guess at the unknowns and test the guesses with simulations. Lee’s post is very much worth the full read, so give it a perusal.

Lee got me thinking with his point that “natural systems don’t have designers.” Evolutionary processes have resulted in the brain we have today, but there was no intention or design behind those process. Our minds are undesigned.

I find that fascinating. In the first place, because it means that simulation will be exceedingly difficult. How do you reverse-engineer something with no engineer? Second, even if a simulation is successful, it by no means a guarantees that we can change the substrate of an existing mind. If the mind is an emergent property of the physical brain, then one can no more move a mind than one could move a hurricane from one system to another. The mind, it may turn out, is fundamentally and essentially related to the substrate in which it is embodied.UPDATE: Hanson’s main claim is that we can make “economically-sufficient substitutes for human workers” at some point because “we already have pretty good signal-processing models of some cell types; we just need to do the same for all the other cell types.”  I acknowledge non-biological minds may be possible, but not within this century.

The problem is two-fold. First, we don’t understand the neurons in sensory organs well enough to replicate them, we just skip over broken ones with an entirely different system and let the brain figure the data out. The mind can “see” with almost any sufficiently nuanced set of information. As Lee notes, we can do that because we already know how to make an artificial ear, the problem was connecting it to a brain. Second, understanding the neuron does not mean we understand the network, nor how the network uses each neuron to process and retain information. Lee’s protein example is designed to show this micro/macro incongruence.

Finally, even if I grant Hanson’s point that the mind was “designed,” evolution is a messy and lazy inventor. As such, replication of biological structures is notoriously difficult. Perhaps unplanned is a better word?

I’m eager to see how this debate progresses, but it seems to me that Hanson takes the computer analogy a bit to strongly.

Image of DTI Sagittal Fibers via Wikipedia

Here’s the extended version of our interview with director Joe Kosinski from the December issue of DISCOVER, in which the first-time feature film director talks about reinventing the light cycle, building suits with on-board power, and how time passes in Tron compared to the real world.

Why return to Tron, and why now?

The original Tron was conceptually so far ahead of its time with this notion of a digital version of yourself in cyberspace. I think people had a hard time relating to in the early 1980s. We’ve caught up to that idea—today it’s kind of second nature.

Visually, Tron it was like nothing else I’d ever seen before: Completely unique. Nothing else looked like it before, and nothing else has looked like it since—you know, hopefully until our movie comes out.

How did you think about representing digital space as a physical place?

Where the first movie tried to use real-world materials to look at digital as possible, my approach has been the opposite: to create a world that felt real and visceral. The world of Tron has evolved [since it's been] sitting isolated, disconnected from the Internet for the last 28 years. And in that time, it had evolved into a world where the simulation has become so realistic that it feels like we took motion picture cameras into this world and shot the thing for real. It has the style and the look of Tron, but it’s executed in a way that you can’t tell what’s real and what’s virtual. I built as many sets as I could. We built physically illuminated suits. The thing I’m most proud of is actually creating a fully digital character, who’s one of the main characters in our movie.

What did you keep from Tron, and what evolved?

The central one in that whole equation is the character of Kevin Flynn, played by Jeff Bridges. For me, that was the first thing I did in getting the project going—securing Jeff’s commitment to be involved, to continue the character of Kevin Flynn to bridge the gap between the original and ours. We brought back Clu, Jeff’s avatar. Kevin F is now 60 years old, where Clu is frozen at 35 years old. But to expand on that, we’ve introduced Sam Flynn, who is Kevin Flynn’s son. He’s really the character the audience meets first, and we follow his journey as he goes back into the computer to look for his father who disappeared 20 years ago.

What else did the original directors want to do, but couldn’t, that you had the ability to include?

One thing that turned out to be special about our film is illuminated suits. All our suits are self-powered and self-illuminated, so you never have to fake light when two characters get near each other. They can illuminate each other’s face and reflect in the floor and the wall around them, or light up a room when you turn the lights out.

Was it a huge hardware challenge to build those suits?

Absolutely. You’re talking about flexible light fabric, and everyone had to carry their own on-board power. It couldn’t be a big bulky spacesuit—these are really elegant, thin suits that you had perform physical stunts in and wear comfortably.

With the advances in computing and filmmaking since 1982, you could do just about anything with Tron: Legacy. How did you keep it grounded?

I didn’t want to make a movie about the Internet. That’s kind of a trap. It can make your movie feel dated the weekend after it comes out. I really liked the idea that this was a closed-off system like the Galapagos Islands, where the simulation has been constantly evolving and growing on some server locked away in some hidden place. That way it felt more like a Western: The world is large and expansive, but at the same time, there’s a code or a set of rules you have to follow. If you want to send a message to someone, you can’t just beam it across cyberspace. You have to get on your light cycle and deliver it in person. You understand that there are repercussions in physicality that make it feel as non-virtual as possible.

To continue the analogy: If this is a digital Galapagos, what is the driver of evolution in the Tron world?

Kevin Flynn created a system that has the ability to evolve on its own. Today you read about these kinds of life simulations, where you program digital organisms that grow and mutate. It’s cutting-edge stuff, but we’re saying Kevin Flynn was such a brilliant, far-ahead thinking guy that he was experimenting in these new types of code that can do self-generation and evolution. Therefore it doesn’t need maintenance and it doesn’t need input from a programmer to change it. In doing so you can create a magical thing, but you can also create something that can get out of control. That’s one of the themes at the heart of our film: Technology can be a very powerful and useful thing, but if not used correctly or if left unchecked, it can turn out to be a very dangerous thing as well.

Given that Kevin Flynn has been inside Tron for 20-plus years, I would like to nerd out with the following question: Does time work the same inside the digital world as outside?

No, it doesn’t. In the first film they weren’t very specific about it, but you got the sense when Kevin Flynn returned to the real world, very little time has passed compared to what he had endured. We’ve gone with a ratio of basically 50 to 1: for every 50 years spent inside the world of Tron, only one year would pass in the real world. If you’re a user, or a human being pulled into that world, you track with what your age would be in the real world—that’s why Kevin Flynn looks 60 years old in our film. But if you’re a program, you remain the age you were when you were created. Therefore, since Kevin Flynn created Clu in 1985, when Flynn was 35 years old, Clu looked exactly like he did then. And in the 25 years since that moment, Clu hasn’t aged at all. Programs are frozen in time.

How do you maintain many of the Tron elements that seem fairly simple now, like the light cycle game, but keep it relevant in 2010?

If you’re talking about light cycles, for instance—that’s something where I was able to sit with Steve Lisberger, the director and creator of Tron, and go back into the archives. On the original light cycle, the rider was meant to be on it and not enclosed in the canopy. But they just couldn’t render that complex geometry, so they had to create that simple shape that covered the rider up.

Knowing that fact, it was fun to go in with my design team and say, “All right, we want an external rider on the light cycle now that we can do anything. How do we maintain the original intent, but evolve the design from that classic shape?” It’s pretty simple—the original is basically a 2D board where all you do is try to cut off the guy in front of you with these right-angle turns. We’ve taken the 90-degree restrictions away so they move like real racing bikes with big sweeping turns. And we’ve taken away the two-dimensionality of the game boards—now it’s a multi-level game board with ramps that carry you up and down.

Did you invent any new games, or just stick with the originals?

We did invent some new ones, but I’m going to keep those to myself for now.

What did Tron get right about the future? And what do you hope to get right?

If you spend some time with Steve and talk about his goals for that first film and then [look at] our film, the one thing I think both touch on—and I think this is a good lesson to be learned in our society today—is the importance of maintaining human connection in an increasingly digital world. That’s a strong element in our film with the connection between father and son. It’s something that we all deal with every day, and struggle with—being able to unplug ourselves and focus on the people around us. It’s easy to get lost in technology as it continues as it invades more and more of our lives.

Images: Disney Enterprises

In the short term, we may be stuck, but in the longer term, quantum money could help solve the problems by providing a secure currency that can be used without resort to a broker.

Physicist Steve Wiesner first proposed the concept of quantum money in 1969. He realized that since quantum states can’t be copied, their existence opens the door to unforgeable money.

Here’s how MIT computer scientist Scott Aaronson explained the principles:

Heisenberg’s famous Uncertainty Principle says you can either measure the position of a particle or its momentum, but not both to unlimited accuracy. One consequence of the Uncertainty Principle is the so-called No-Cloning Theorem: there can be no “subatomic Xerox machine” that takes an unknown particle, and spits out two particles with exactly the same position and momentum as the original one (except, say, that one particle is two inches to the left). For if such a machine existed, then we could determine both the position and momentum of the original particle—by measuring the position of one “Xerox copy” and the momentum of the other copy. But that would violate the Uncertainty Principle.

…Besides an ordinary serial number, each dollar bill would contain (say) a few hundred photons, which the central bank “polarized” in random directions when it issued the bill. (Let’s leave the engineering details to later!) The bank, in a massive database, remembers the polarization of every photon on every bill ever issued. If you ever want to verify that a bill is genuine, you just take it to the bank”

At this point I should mention that this stuff is all bark and no bite —- theory has vastly exceeded the actual engineering of quantum computers or quantum much of anything engineered, so when I said long term, I meant decades, at least.

But that doesn’t make it any less interesting to ponder. So, Aaronson’s description of quantum money leaves us with same middleman problem. We need a way to check the bill’s authenticity without having to bring it to the bank or any other central institution, and yet still have it be difficult to forge.

As it happens, we have a solution to a version of this problem in today’s secure online transactions. In principle we rely on the difficulty of factoring. It’s easy to multiply two numbers and get a third number. If the third number is really large, it’s computationally time consuming to factor it and find the original two numbers — unless you already have one of the numbers. Using prime numbers as public and private keys, we can send secure transactions all over the Internet, and as long as our computational power does grow exponentially, we can feel relatively secure.

Quantum computing uses a similarly computationally difficult trick, thanks to knot theory. Edward Farhi, David Gosset, Avinatan Hassidim, Andrew Lutomirski, and Peter Shor* published a paper on this technique in 2009, and Technology Review summarized it well:

“Their quantum cash is based on a new kind of asymmetry: that two identical knots can look entirely different. So while it may be easy to make either knot, it is hard to find a way to transform one into the other.”

(Paper authors) Farhi and co. say: “The purported security of our quantum money scheme is based on the assumption that given two different looking but equivalent knots, it is difficult to explicitly find a transformation that takes one to the other.”

Under the proposal, a bank could mint money with a serial number and a partially recorded quantum state. A merchant with a quantum computer could check the money by applying an algorithm using knot theory that went looking for a mathematically identical knot. If the states and the knots match, the money can be accepted.

A friend of mine, who happens to be a knot theory mathematician and who sent me the Farhi paper in the first place, made a good point over email that the authors don’t discuss exactly how we’re going to pass this quantum money around. They mention that paper money could be minted, but how will we pay quantum money over the quantum Internet?  Well, a question for future research, I imagine.

For whatever reason, few sci-fi creators have much to say on the subject of money in the future. Usually they’ve gone to a paperless currency (which does seem like the end point of current trends), and they have some kind of charge card or an account that can be accessed with a biometric security protocol. Beyond that, money is just assumed.  But this always left me unnerved because it left control of money in the Visas and MasterCards of the world.

But with quantum money? We take back our dollars.

*Shor is well known for Shor’s Algorithm, which showed that a quantum computer could factor numbers much faster than a classical computer.

Independence Day has one of my most favorite hero duos of all time: Will Smith and Jeff Goldblum. Brawn and brains, flyboy and nerd, working together to take out the baddies. It all comes down to one flash of insight on behalf of a drunk Goldblum after being chastised by his father. Cliché eureka! moments like Goldblum’s realization that he can give the mothership a “cold” are great until you realize one thing: if Goldblum hadn’t been as smart as he was, the movie would have ended much differently. No one in the film was even close to figuring out how to defeat the aliens. Will Smith was in a distant second place and he had only discovered that they are vulnerable to face punches. The hillbilly who flew his jet fighter into the alien destruct-o-beam doesn’t count, because he needed a force-field-free spaceship for his trick to work. If Jeff Goldblum hadn’t been a super-genius, humanity would have been annihilated.

Every apocalyptic film seems to trade on the idea that there will be some lone super-genius to figure out the problem. In The Day The Earth Stood Still (both versions) Professor Barnhardt manages to convince Klaatu to give humanity a second look. Cleese’s version of the character had a particularly moving “this is our moment” speech. Though it’s eventually the love between a mother and child that triggers Klaatu’s mercy, Barnhardt is the one who opens Klaatu to the possibility. Over and over we see the lone super-genius helping to save the world.

Shouldn’t we want, oh, I don’t know, at least more than one super-genius per global catastrophe? I’d like to think so. And where might we get some more geniuses? you may ask. We make them.

In his essay, “The Singularity: A Philosophical Analysis”, philosopher David Chalmers notes that there is a very real chance that if machines become self-aware and start improving themselves, we’re going to have a problem (*cough* Skynet *cough* Liquid T-1000 *cough, cough*). One of his potential solutions is to enhance ourselves to keep up:

This might be done genetically, pharmacologically, surgically, or even educationally. It might be done through implantation of new computational mechanisms in the brain, either replacing or extending existing brain mechanisms. Or it might be done simply by embedding the brain in an ever more sophisticated environment, producing an “extended mind” whose capacities far exceed that of an unextended brain.

Does any of that sound familiar? Perhaps a little film called Gattaca may ring some bells? Chalmers is arguing enhancement may be necessary to prevent extinction. Why not extrapolate that logic to other existential risks. Alien invasion? Superhumans would probably put up a better fight. Skynet goes live? An army of hackers with a collective IQ of 200+ and neuro-integrated interfaces would clean that up in a jiffy. But what about our current problems? Although heavy-handed, the message in both versions of The Day the Earth Stood Still is that humanity’s greatest existential threat is itself. War, suffering, poverty, and environmental destruction all seem like problems that would merit allowing our best and brightest to become even better and brighter for the sake of everyone.

A common fear is that the super-intelligent would just step on us normals, creating second-class citizens. Enhancement doesn’t just mean the ability to do complex equations and create new molecular compounds; raw intellectual horsepower is just one among many possibilities. We know that some people have moral problems caused by damage to specific parts of their brain. As neuroscience progresses, there is a very real possibility we’ll be able to improve those specific parts of the moral brain. I don’t mean we’d have a society of lock-step rule followers, but instead people who were genuinely better at being moral than most of us. Can you imagine a world where politicians had improved ethical scruples? Or, to put it simply, where the most brilliant minds were also the most caring?

Which brings me back to Jeff Goldblum in Independence Day. Not only does he come up with the solution, but he selflessly gets in the nuke-strapped UFO with Will Smith to fly into the middle of the enemy mothership. Same for professor Barnhardt, who is as good at moral philosophy as it seems he is math, attempting to show Klaatu the best of our species.

In science fiction, when humanity is faced with existential crises, we turn to great minds attached to great hearts. While we aren’t under alien attack or facing sentient machines, our world has its own share of problems. Human cognitive enhancement might just be the solution from which all other solutions are born; or maybe it brings too many risks of its own.

ID4 Promotional Image via Wikipedia under fair use

The full audio of the event can be streamed or downloaded from here.

We all pitched in to comment on the clip featuring Keanu Reeves learning kung fu through an apparently painful download in The Matrix. The panel consensus: if something like a neuroprosthetic arm for everyone is in the near future, downloading skills a la The Matrix is at the far end of the far future. Reasoning: there are hundreds of thousands of sensory and movement neural channels being activated while learning of kung fu (not even counting vision, which has a million channels per eye). To train the brain via download, we’d either need to excite those channels in just the same way artificially — at roughly normal speed — or figure out how to directly modify the many millions to billions of neurons in the brain that are changed while learning kung fu. Either option presents technical challenges we are far from overcoming.

I picked the Minority Report clip, which featured robotic spiders artfully killing any last doubts you might have had of having privacy in the future. In this clip, some police come to an apartment complex that they are searching for a person in, and release a platoon of nimble robot spiders. These spiders spread out and crawl up people to scan their retinas to identify each person in the building. They sense in the infrared (which is why Tom Cruise hides in a tub of cold water) to detect the warmth of live bodies to be scanned. One of the brilliant aspects of the way it’s shot, as a pan over top the exposed rooms of a floor of the building, is how it shows just how “normalized” the loss of privacy has become in the future, with one couple in the midst of a fight hardly pausing their exchange of blows to let the scan happen before starting to whale at each other again. It’s as natural as selling a row of pumpkins on FarmVille and losing your privacy through Facebook application data misuse.

There’s a few things I love about this segment of the film. The first is that, like most good sci-fi, it simultaneously makes you say “oh wow that’s cool,” while terrifying the crap out of you that this may be the endpoint of all the privacy failures we are being subjected to. Sci-fi as incubator of dreams and place to work out our anxieties about technology.  On a professional level, I also liked how center stage was not a humanoid robot for once, but rather a non-human biologically-inspired robot. I appreciate that story-tellers need robots that people can relate to, but the disconnect between what actually goes on in robotics (where humanoid robotics is a tiny fraction of research effort) and what’s always in the movies is sometimes jarring. Not only did Minority Report show a biologically-inspired robot, it showed them in exactly the context in which they make a lot of sense: solving problems that conventional machines and robots don’t do well, such as high agility motion that needs large amounts of sensory intelligence. Animals are fantastically agile. But agility requires a lot of flexibility in the way a body can move, and with that flexibility comes the great challenge of how to control all that movement for stable motion, and how to acquire enough sensory information to guide the body in a highly nimble way. It’s a fantastically complicated problem, and understanding how it works is precisely what motivates some of us who do research in this area.

I also liked how the makers of the movie went to the trouble to seek out a colleague who studies jumping spiders, Damian Elias at UC Berkeley, to get good sound of the spiders scampering around.

It’s interesting, as Sean Carroll noted for a similar panel he was part of  at Comic Con, how much demand there is for this kind of discussion. With the blogosphere and traditional media saturation of science and tech news, maybe this all portends the dawning of a new age of sci-fi for viewers who will be a lot more sophisticated in the kinds of stories that will get them intrigued.

CLARICE: Zoe Graystone was Lacy’s best friend. A real tragedy for all of us. She was very special. I mean, she was brilliant.

NESTOR: At computer stuff, right? That’s my major. Did you know that there are bits of software that you use every day that were written decades ago?

LACY: Is that true? Oh, that’s amazing.

NESTOR: Yeah. You write a great program, and, you know, it can outlive you. It’s like a work of art, you know? Maybe Zoe was an artist. Maybe her work… Will live on.

From: Rebirth, Season 1.0 of Caprica

I’m excited that today Caprica is back on the air for the second half of its first season. As the show’s science advisor, I thought I’d pay homage to its reentry into our living rooms with some thoughts about how the show is dealing with the clash between the mortality of its living characters and the immortality of its virtual characters.

As Nestor says in the passage above, a great program can outlive you. It’s a clever reference to Zoe’s hacking leading to her digital self outliving her biological self. But the extent to which a piece of art can outlive you differs radically depending on what you “paint” with. Philosopher John Haugeland’s 1981 work “Analog and Analog” is full of great insights into the difference between digital and analog systems. Haugeland defines a digital device as*:

1. A set of types (for example, our alphabet; or 1 and 0s of computers),

2. A set of feasible procedures for writing and reading tokens of those types, and

3. A specification of suitable operating conditions, such that

4. under those conditions, the procedures for the write-read cycle are positive and reliable.

Using Haugeland’s criteria, if you are dealing with words, then you are dealing with a digital system. Shakespeare’s sonnets are largely as he wrote them, and they will remain this way for an eternity, perhaps with small adjustments to account for changes in English (though there will always be versions in the original form for scholars). Because the alphabet is a set of types (1), and we have a set of feasible procedures for writing and reading tokens of those types (2), and a specification of suitable operating conditions (e.g., light to read and write by!) (3), such that under those conditions, we can read the sonnets and copy them in a reliable fashion without error (4). So the work of Shakespeare will live on, potentially eternally, because it is enmeshed in digital system (written language) in which perfect copying is possible. We don’t complain, when we read a reprint of his work, that we are not “reading the original”—it is the original.

Now contrast this with a Rembrandt painting, which can never be perfectly copied; so, once—despite the efforts of art conservationists—it finally turns to dust, this Rembrandt painting will be no more. Eventually all of his work will be lost this way. Clearly, a Rembrandt does not fit the criteria for a digital device.

An interesting analog to these concepts is the idea that biological aging represents the accumulation of small errors of copying DNA over the course of your lifetime. If the copying was perfect, again we’d be a digital device—and eternal.

In Caprica, Zoe’s and Tamara’s switch from their biological to virtual selves leaves a lot for their biological family and friends to ponder, similar to questions that characters in BSG had about the 12 human-looking cylon models. Are these virtual selves “real”? Do they have a “soul”? How does the lack of aging, and impossibility of death, affect relationships between digital and analog selves? How can a digital self inhabit both the virtual worlds of Daniel Graystone’s creation and the analog world? These are some questions that will continue to lurk beneath the surface as we see the stories of series 1.5 unfold. Stay tuned!

Reference: Haugeland, J. (1981). Analog and Analog. Philosophical Topics, 12, 213-226.

Below are short excerpts of the guest scientists’ responses, with links to the full versions:

Ken Caldeira: “…If you could directly produce chemical fuel from sunlight and do it affordably, that could really be a game changer…”

Jack Horner: “…If we want to see an animal like a velociraptor, we will be able to create one by genetic engineering. It might even be possible to make something that looks like a T. rex…”

Oliver Sacks: “…We thought that every part of the brain was predetermined genetically, and that was that. Now we know that enormous changes of function are possible…”

Sylvia Earle: “…We’ve explored only about 5 percent of the ocean. For us to have better maps of the moon, Mars, and Jupiter than of our own ocean floor is baffling…”

Rodney Brooks: “…The arguments we have about drugs and sports are minuscule compared with what’s coming, such as ‘What is the definition of human?’ We have the Paralympics now, but we’ll have the Augmented Olympics in the future…”

Debra Fischer: “…Every year since 1995, we have discovered more extrasolar planets than the year before. A parallel thing could happen with extraterrestrial life: After we find one example, we’ll hone our strategies to be smarter and more efficient…”

Tachi Yamada: “…I don’t believe just because you’re poor, you shouldn’t have access to lifesaving technology…”

Neil Turok: “…The science has reached the point where questions that used to be just philosophy could be observationally testable in 10 or 20 years…”

Ian Wilmut: “…We should be able to control degenerative disorders like Parkinson’s and heart disease…”

Sherry Turkle: “…Sometimes a citizenry should not ‘be good.’ You have to leave room for real dissent…”

Brian Greene: “…We may establish that there is not a unique universe—that ours is just one of many in a grand multiverse. That would be one of the most profound revolutions in thinking we have ever sustained…”

These artificial intelligences, not likely to have had the “nostalgia module” installed, may quickly flee the home planet like a teenager trying to pretend it isn’t related to its parents. If nothing else, they will likely need to do this to find further resources such as materials and energy. Where would they want to go? Shostak speculates they may go to places where large amounts of energy can be obtained, such as near large stars or black holes.

Stephen Hawking imagines aliens covering stars with mirrors
to generate enough power for worm holes

Stephen Hawking has suggested one reason to go to high-energy regions would be to make worm holes through space-time to travel vast distances quickly. These areas are not hospitable to life as we know it, and so are not currently the target of SETI’s telescopes searching for signals of such life.

In the same article, Shostak also makes the argument that since biological intelligence is a short stepping stone to artificial intelligence, “the majority of the intelligence in the universe could well be artificial intelligence.” There’s clearly a missing premise here, which is that biological intelligence means an intelligence that invents radio or TV, or more broadly speaking, technology. But this is clearly false. From cuttlefish to corvids, the scientific evidence for high levels of intelligence in non-human animals is rapidly accumulating. At the moment, it’s not even clear that the invention of technology will be good for us as a species: an analysis of nine planetary boundaries within which human life can flourish shows that we are now transgressing three of these. Given that life has flourished for billions of years, for this to happen with just a few thousand years of agriculture and a few hundred years of industrialization shows that the step from advanced technology to artificially intelligent descendants roaming the galaxies is not one to be taken for granted.

In any event, given we can’t look everywhere, should thoughts about AI inform where we look? I don’t think so. First, based on our very limited experience, only Homo sapiens, just one of tens of millions of species of life on Earth, have developed technology. Were it not for our species, it’s unclear whether technology would ever have come about on Earth. Second, it’s far from obvious that our species will have the maturity to survive the power of our achievements for more than a blink of evolutionary time–the development of AI that leaves this planet, or at the very least serious efforts toward space colonies, is probably our best hope for long term survival–but we may not get there. Perhaps the situation is no different for other forms of life that have developed technology. They will have all emerged from a Darwinian primordial soup, a soup where certain vicious and short-sighted traits will have been essential to survival. Third, it would probably be both more successful and more scientifically useful to adjust our search strategy to improve the chances for finding extraterrestrial life, rather than intelligence.

My personal favorite for such a tweak to our search strategy is to look for places that have the hallmarks of increasing entropy. All forms of life take in energy that has some degree of entropy and re-emits it with increased entropy, such as heat. For our biosphere, we absorb sunlight and reflect heat, which appears as a “red edge” in the spectrum of reflected energy. The same, incidentally, seems likely to be true of artificial intelligence: it will require energy such as electric power, which will be radiated at higher entropy, such as the heat of integrated circuits. Sean Carroll has written an excellent explanation of the red edge in one of his postings over at Cosmic Variance. If we build better red edge detectors, we will both improve our chances of finding the much more common non-technologically savvy forms of life in the universe, and as an added side benefit, we might just detect the much rarer roaming AIs out there — although, as Hawking suggests, we may want to avoid hailing them down for coffee.

Image from Stephen Hawking’s Universe, “Fear the Aliens”

Having robots with minds implementing the scientific process rather than math problem solving is essentially what’s happening in a few corners of robotics, most recently with the Xpero project, an effort to develop an embodied cognitive system that learns about its world much like an infant would. It’s one of a host of robo-infants being worked on (here’s a nice overview graphic). This approach has led to some very impressive achievements including an “evil starfish” robot that can quickly learn how to control its body after several of its “limbs” have been chopped off.

Hod Lipson (left) and yours truly pulling legs off the evil starfish in 2006.

In 2006, Hod Lipson and co-workers published a short paper with the sexy title “Resilient Machines Through Continuous Self-Modeling.” In it, he demonstrated how a small, starfish-like robot (aka, “the evil starfish”) could automatically learn its own body shape and movement capabilities. It did this through an automatic process of scientific inquiry. It worked something like this: first, make an arbitrary movement. What this means is that the robot sends out signals to its body, without knowing what those signals will do. While sending these movement signals out, the robot records sensory signals that tell it about what happened to the body due to that movement (scientific process analog: experiment). Second, generate a small number of models of the body that are compatible with movements resulting in the recorded sensory information (analog: hypothesis generation). Third, through some fast on-board simulation (aka, thinking), the robot figures out what movement(s) would give it the most information to distinguish between the different body models that are compatible with the information it has collected (analog: prioritizing hypotheses for testing). Fourth, the robot executes these movements, and uses the resulting sensory information for further refinement of its guess as to what its body is (analog: hypothesis testing and refinement).

What is great about this process, as I discovered when I visited Lipson’s lab some years ago to give a talk at Cornell, is that the robot has an amazing degree of robustness. The starfish robot shown in the photo has had one of its arms pulled off, and after a brief learning process, it figures out its new body shape and saunters off! It was slightly unnerving to witness this process. There is something about an animal recovering from damage that gives us a sense that it cares about its continued existence. In some sense, this is part of the essence of what it means to be a living organism: something that cares about its continued existence and acts so as to further that goal. When you see a machine act in this manner, it triggers certain associations that make it feel biological.

If indeed, as Alison Gopnik and others have argued, we all grow up absorbing all the important things we need to know through something like the scientific process, then the current work on making an algorithm that emulates the scientific process may be just the thing that AI needs for making breakthroughs on solving the problems we really want our robots to solve, such as making us a coconut mojito with just the right amount of muddled mint.

For more information on the European Xpero project, visit their website. A prior project, also EU-sponsored, was the iCub. A nice overview graphic of different robot infant approaches was in this July’s issue of IEEE Spectrum. Some interesting recent work on formalizing the discovery of regularities through experiments can be found in Hod Lipson’s “Selected Recent Publications.” Here is a thoughtful commentary on the starfish robot work by Chris Adami. Using data to automatically do science has also received attention in bioinformatics, most recently highlighted in articles about Sergey Brin’s datamining efforts to find a cure for Parkinson’s, in this podcast, and in academic circles here and here.

Image of robot in crib by Malcolm MacIver using free 3-D models on TurboSquid.

My complaint isn’t that he has set a date by which we’ll understand the brain, but that he has provided no baseline value for his exponential growth claim, and has no way to measure how much we know now, how much we need to know, and how rapidly we will acquire that knowledge.

The part which I have bolded is the central flaw of much Singul-itarian thought. They cannot show how technologies are changing exponentially or explain why that will allow whole brain emulation. Even if Kurzweil could give us the necessary data, exponential growth does not guarantee practical results, as I discussed in my post on progress in genomic sequencing. George Dvorsky, a Canadian futurist and colleague of mine at the Institute for Ethics and Emerging Technologies, defends whole brain emulation as a possibility, but hedges his bets on the timeframe:

Kurzweil’s prediction of 2030 is uncomfortably short in my opinion; his analogies to the human genome project are unsatisfying. This is a project of much greater magnitude, not to mention that we’re still likely heading down some blind alleys.

My own feeling is that we’ll likely be able to emulate the human brain in about 50 to 75 years. I will admit that I’m pulling this figure out of my butt as I really have no idea. It’s more a feeling than a scientifically-backed estimate.

Since we’re pulling figures out of our posterior, I’m going to throw a guess from a futurist I really respect, Gene Roddenberry, into the ring: successful whole brain emulation will first occur in the twenty-fourth century. That’s when Data and his unique “positronic brain” was built in the Star Trek universe. Seems like a more reasonable time frame to me.

Why? Consider the state of current prosthetics and robotics. The most advanced robotic arms in the world, be they the iLimb or DEKA’s Luke Arm, are cumbersome, heavy, weak, clumsy, and delicate compared to a human arm. They lack touch receptors; texture, heat, density, and other basics of sensation are decades away. One of the only successful osso-integrated prosthetics is a set of what are essentially peg-legs of metal and rubber on a cat named Oscar. Don’t even get started with me on neuro-integration. In short, we’re still struggling with replicating knees and elbows, so how the hell is the whole brain a mere few decades away?

The answer to the titular question of this post would seem to be: nope. At least Kurzweil is adhering to the Law of Futurology!

Image by digitalbob8 via Flickr

While I was still a doctoral student, I had the opportunity work with a relative of interactive holographics, 3D virtual reality data CAVEs. This particular one, at the National Center for Supercomputing Applications (NCSA) in Urbana Illinois (the birthplace of HAL) circa 1999, was a cube with back projection on five of the six walls. You wore a headset that tracked your head position and orientation, and goggles that were LCD screens that blocked images to your right eye when the projectors were rendering images for your left eye, and vice versa when the projector was displaying images for your right eye. As you walk through space or move your head, what you see in the virtual space changes as you would expect it to.

The problem that had pushed me to use this system was trying to analyze 3D motion data of a fish that I was conducting research on. I’d developed a motion capture system for the fish, which gave fantastic 3D data of the fish moving while it was attacking its prey, but looking at this 3D data on 2D computer monitors turned out to be quite difficult. Even replaying the motion from several different views didn’t quite do the trick. So Stuart Levy at NCSA put my data set into a system called “Virtual Director” and I was able to playback the data in the cave. It was something of an unbelievable experience the first time I tried it – suddenly I could walk around the animal as it engaged in its behavior, manipulate it to get any view, rotate the wand I held to wind the behavior forward or back at different speeds. Visitors particularly enjoyed my “Book of Jonah” demo where I positioned them so that they ended going into the mouth of the fish during a capture sequence.

For my technical problem, the VR CAVE was appropriate technology: 3D display and interaction for an inherently 3D data set. It helped me see patterns in the data that I had not clearly seen before, which were incorporated into some of my subsequent publications that analyzed the movement data. It was worth the effort, and the physicality of it was fine since I didn’t need to spend multiple days working through the data.

Other uses of these kinds of “direct manipulation” interfaces that mix 3D data and real world interaction have not found such a receptive audience, as people complain that it seems tiring to make sweeping (if dramatic) gestures to go through photos that would just as well be navigated through with an arrow key. As someone who still uses “vi” to edit my text with, I can relate to criticisms of interfaces that offer more than is needed.

The important question, for any given interface, is whether simplifies difficult problems of control or analysis, or gets in the way. My former colleague Don Norman at Northwestern University has contributed a great deal to our understanding of this question, in books like The Design of Everyday Things. One of my favorite examples from that book considers two different interfaces to manipulating the position of a car seat. In one interface, on a luxury American car, there is a panel of knobs and buttons almost hidden below the left side of the dashboard. To go from a state of discomfort to a new chair position requires translating your discomfort into a series of knob pulls and twists on a console of many controls with tiny labels below each. In contrast, a German luxury car had a small version of the driver’s chair in the dashboard. To move the back of your chair down, you manipulated the chair in the dashboard accordingly; to move it forward, you would move it in the direction the chair was facing, and so on. One interface placed a large cognitive load on the user to solve the discomfort problem, while the other placed minimal demands.

Another favorite example is the “speed bug” – a tab that a plane pilot puts on the edge of an airspeed indicator to mark the velocities for critical changes to shape of the wing. Were it not for those bugs, the pilot would have to remember the velocity to do the wing adjustments – and that’s not easy, because it changes with things like the weight of the plane.

The virtual fish, miniature car seat adjuster, and speed bug are all examples of interfaces that make problems easier, and in this sense, amplify our brain power. Interactive holographic interfaces can do the same for problems where space is a convenient or needed basis for navigating the information. This isn’t always apparent in sci-fi depictions of these interfaces, but their use speaks to our hope that such 3D holographic wizardry will help us cope with the flood of data we contend with on a daily basis.

Doctors are not doing so well. In addition to being extremely expensive to train, maintain, and, of course, to visit, they have a lot of other problems. If your doctor is a drunk, an addict, or just plain-old incompetent, his or her colleagues may not tell you or anyone else.[1] Even when doctors are sober and sharp, their diagnoses are often, ahem, less than correct. Mark Walker’s “Uninsured, Heal Thyself” paints a pretty terrifying picture:

Physicians can and do misdiagnose frequently: they prescribe for nonexistent diseases or injuries and fail to notice symptoms or make the correct inferences. An article in the Journal of the American Medical Association noted: “Two 1998 studies validate the continued truth that there is an approximately 40% discordance between what clinical physicians diagnose as causes of death antemortem and what the postmortem diagnoses are” (Lunberg, 1998). This is a pretty shocking statistic: in 4 out of 10 deaths there is a disagreement between what physicians think is the cause of death prior to autopsy, and autopsy findings.[2]

Egads. Is there any solution to the doctor debacle? Walker proposes computer-aided diagnosis:

For example, in a well-known 1971 study, a computer diagnostic system was pitted against experienced physicians in the diagnosis of acute abdominal pain: computer diagnosis was 91.1% accurate compared to 79.7% for experienced physicians (de Dombal et. al., 1972). In another study, computer diagnosis matched that of neurosurgeons, orthopedic surgeons and general practitioners in overall average in diagnosing lower back pain. While humans surpassed computers in non-critical cases, computers surpassed humans in diagnosing more critical spinal symptoms in which quick intervention is correlated with better outcomes (Bounds et. al., 1998).

The X-Prize Foundation (famous for spurring the privatization of space flight) agrees with Walker. The foundation is developing a new prize: the “AI Physician X Prize, which will be won by the first team to build an artificial intelligence system that can offer a medical diagnosis as good as or better than a diagnosis from a group of 10 board-certified doctors.”[3] Ten doctors – forget a second opinion, every diagnosis would come with a tenth opinion!

And where would one keep such an artificial intelligence? Why in a smartphone, of course. A hand-held computer used to diagnose medical issues: that sounds suspiciously like something Dr. Bev Crusher might be using on the USS Enterprise; namely a tricorder. Or, as Dilbert creator Scott Adams calls it, an exobrain.[4] My iPhone already has access to Wikipedia, WebMD, the Mayo Clinic and a free app for the University of Maryland Medical System’s medical encyclopedia. In a pinch I could probably use it to help in an emergency until professionals arrived. My knowledge and intelligence is expanded instantly by virtue of owning a hand-held computer with a wireless data signal.

Now imagine an app as smart and accurate as a panel of ten doctors in the hands of a trained MD or EMT, emphasis on the “trained.” Walker’s essay focuses on allowing patients to self-diagnose, but the huge benefit would be for professional diagnoses. Instead of being required to memorize thousands of potential diseases and syndromes, each with their own fickle and bizarre permutations, a doctor’s two primary goals would become 1) ensuring accurate, exhaustive entry of symptoms into the tricorder and 2) giving comprehensive, patient oriented care. Diagnoses, particularly esoteric ones, would become the prerogative of the device, instead of certain hobbled, cantankerous MDs named “House.” In addition to the symptoms entered by the doctor, the tricorder would have access to the patient’s entire medical history — including reoccurring issues, worsening conditions, potential genetic dispositions, and a plethora of other minutia — that could be the difference between sending someone home with “drink fluids and come back if it gets worse” and hospitalization. Furthermore, long, infection-prone hospital stays for “observation” would be reduced or even eliminated thanks to better initial diagnoses.

With so much potential to help, the tricorder may become an ever-present part of the doctor’s uniform, just as the stethoscope did in a previous era.

1. “Doctors don’t rat out their incompetent colleagues,” Salon
2. “Uninsured, Heal Thyself” JET Press
3. “The Next Five Years of the X-Prize” CNET
4. “Exobrain” Dilbert Blog

These are remarkable effects with many potential implications, and applications (next time you’re trying to sell something, make sure you’re seated in a hard chair, and your buyer is in soft chair, for example; and clothes designers have a whole new dimension to consider). What is their underlying basis? The researchers hypothesize that our experiences with touch early in our development provides a scaffold for the development of conceptual knowledge. In adult life, these same touch experiences activate the scaffold in the same way, and lead to unconscious influences on our attitudes and decision making. The experience of weight gets metaphorically associated with seriousness and importance. Idioms like “that’s heavy” reflect this association. Similarly, rough textures get associated with difficulty, and we say “having a rough day.”

This research is another example of how the way we think is all wrapped up in the way we body. The new results add to our growing understanding of the ways in which embodiment and thought are more intertwined than was previously believed. The ways in which cognition is embodied was also the topic of a recent volume in the Cambridge Handbook series, called “The Cambridge Handbook of Situated Cognition”, which I had the pleasure of writing a chapter for.

Research into the ways in which cognition is intertwined with bodily experiences raise interesting issues regarding common science fiction fables and science fact predictions. Many of these hinge on being able to dispense with the body. The body, in this view, is just a convenient output device, easily replaced with another, or not replaced at all so as to be a disembodied intelligence like Hal of 2001. Our body is the computer, and who we are is the software, so if the hardware falls short we can just get new hardware. But what if who we are is this particular software running on this particular kind of hardware? The Cylons of Battlestar Galactica are an interesting mix of these ideas. They never died: as soon as their current body was eliminated, their consciousness was uploaded to another body. But they were not into body swapping: you got uploaded to the same body model, or not at all (aka, death), potentially compatible with embodiment ideas.

St. John the Baptist wearing a hairshirt, by Jacopo del Sellaio, 1485.

Number Six of Battlestar Galactica, a Cylon.

There are many problems with this idea, but despite these problems, it still remains a compelling benchmark, and one that has yet to be reached. But think of the following variation: rather than have your computer and human team answer any old question, the questions have to be similar to what you would expect on the quiz TV show Jeopardy! – clues about trivia in the form of answers to a question that you must come up with.

Even this greatly restricted version of the Turing test is very challenging, but I.B.M.’s machine called “Watson” has recently made intriguing steps toward passing it. Watson takes any Jeopardy-type question and gives a response. It was not developed as a new type of intelligence test, but instead as a grand challenge to beat a human at a language-based task, like a Deep Blue of language (IBM’s Deep Blue chess playing computer beat the world chess champion in 1997). You can challenge it yourself here. It currently uses a fixed set of a large number (in the millions) of documents and a sophisticated parallelized statistical algorithm running on a supercomputer. By being parallelized, the algorithm can try a large number of possible interpretations of the question out at once, and pick the most likely interpretation.

One of the problems with the original Turing test, which the Jeopardy test also suffers from, is that only forms of intelligence expressible through language are eligible for testing. It’s perfectly consistent, for example, for a robot to pass the Turing test, but then fall over when it tries to take its first step. Walking is a skill. It’s learned by a different part of the brain than are facts like what a neti pot is. Memory for these skills is called “procedural memory,” while “declarative memory” is the kind of memory for facts – and many studies within neuroscience have shown that these two forms of memory live in different places in the brain.

But just as we can identify what sorts of intelligence would fall through the cracks of the Turing test, it’s also interesting to think about the kinds of intelligence a Jeopardy test could test for, and where it might fail relative to the Turing test.

For instance, while a machine that passes the Turing test should convincingly answer questions about its first kiss, Watson would be pretty stumped on that one. Presumably, a machine that convincingly describes what their first kiss was like is being masterfully deceptive (at least for the first few models!) – but nonetheless, describing “first-person” states like the emotions that went with your first kiss is quite complex, and faking this convincingly is hard. In this way, we can see the Jeopardy test is a less demanding test than the Turing test. But that’s good – the Jeopardy test is a big challenge as it stands and it will be a remarkable breakthrough if Watson ultimately succeeds when it goes head to head against a human, as it is expected to sometime this fall. Having more attainable tests leading up to the Turing test is very helpful to researchers in artificial intelligence.

While we know not to expect Watson to get the answers to questions regarding feelings right, there are also some simple facts we can’t expect it to know. For instance, a computer that passes the Turing test must be able to give convincing answers to random biographical questions like “where did you get your first report card?”, but giving Watson the Jeopardy form of this question — “where I got my first report card” — should stop him in his tracks.

Can you think of other examples of things that would pass the Turing test but leave Watson with smoke coming out of its ears? Leave your thoughts in the comments.

Moore’s Law (the speed of computing per dollar doubles every 18 months), perhaps the representative concept of ever accelerating technological progress and the foundation of Ray Kurzweil’s prediction of a technological singularity, looks pretty pathetic. And genomics is still getting cheaper and faster:

The genome sequenced by the International Human Genome Sequencing Consortium (actually a composite from several individuals) took 13 years and cost $3 billion. Now, using the latest sequencers from Illumina, of San Diego, California, a human genome can be read in eight days at a cost of about $10,000. Nor is that the end of the story. Another Californian firm, Pacific Biosciences, of Menlo Park, has a technology that can read genomes from single DNA molecules. It thinks that in three years’ time this will be able to map a human genome in 15 minutes for less than $1,000. And a rival technology being developed in Britain by Oxford Nanopore Technologies aspires to similar speeds and cost.

In terms of reductions in cost and speed, genomic sequencing has accomplished in a decade what took computing has been lurching towards for over half a century. Imagine going from the ENIAC to the iPad in fifteen years instead of sixty—that’s what genomic sequencing did. It’s hard to even imagine what will happen when a person’s genome can be sequenced in under an hour for the cost of a new computer. On its own, this news is amazing. Genomic sequencing’s progress is also frustrating, however, given the near non-existence of drugs and treatments based in genomic research. Only now, with the ability to sequence multiple genomes at a reasonable (if still rather steep) price, are researchers beginning to gain traction. While it’s well and good to be able to sequence a genome, we’re still a ways off from doing anything with the information.

The steep fall of costs and time for genomic sequencing mixed with the significant lack of real, medical applications like drugs or diagnostic tools has troubling implications for those committed to the futuristic prophesies of a technological Singularity. The technological Singularity, pushed by Kurzweil and his zealots at the Singularity University (recently profiled in the New York Times and rightly ridiculed at Scientific American) is based in an extrapolation of Moore’s Law to all of human technological progress. Kurzweil believes that, based on his calculations, technological change is accelerating at such a pace that by 2045 progress will become incomprehensibly rapid. Things like genuine artificial intelligence, self-replicating nanorobots, and human-machine hybrids will be effortless and prolific. Yet, in less than a decade, genomics has shown that improvements in the cost and speed of a technology do not guarantee real-world applications or immediate paradigm shifts in how we live our day-to-day lives.

Moore’s Law is moot and genetics (in terms of applications) is just beginning to boom. Back to the drawing boards, futurists: acceleration alone isn’t enough to build the future.

(image: The Economist)

Which is exactly what the students and scientists at the Institute for Dynamics and Systems Control in Switzerland pulled off at the end of the semester last year, when they created the Distributed Flight Array. The devices they engineered look like hexagons made of white plastic, each with a propeller in the center. Alone, each device is autonomous, but  pretty dumb, mostly just wandering around the floor and occasionally lifting into unstable flight. But as each device bumps into another, they dock. When they reach a critical number, the collective becomes much greater than the sum of its parts.

What’s clever here is not creating a flying device, but using multiple small intelligences to combine into one larger intelligence. From the DFA website:

Joined together, however, these relatively simple modules evolve into a sophisticated multi-propeller system capable of coordinated flight. The task of keeping the array in level flight is distributed across the network of vehicles. Vehicles exchange information and combine this information with their own sensor measurements to determine how much thrust is needed for the array to take-off and maintain level flight.

The devices have no immediate practical purpose. Raymong Oung, a staffer at the institute, said in the YouTube comments that the device is “a research and teaching tool for distributed estimation and control.” Still, it’s not hard to imagine finding practical uses for a device that can roam around an area gathering information, join together with similiar devices, and then fly off to somewhere else.

* By the way, Mattel just re-acquired the Voltron license, and there are new Voltron cartoons coming on Nickelodeon. Gen X, start your engines!

Consultants like Newssift, ScoutLabs, and Jodange use complex algorithms to scan keywords in remarks about corporations made on Twitter and Facebook, then categorize them as positive or negative via filters—the companies say they can even parse sarcasm, slang, and other linguistic nuances. Filters can sift through levels of positivity/negativity, intensity. Some can also identify more influential opinions from those social-media hubs and tastemakers. As the tech becomes increasingly sophisticated, it may become more prevalent in standard search engines or predict future developments like stock price fluctuations.

Companies are interested in measure online opinion, of course, because the perception of the company or its products can have a strong effect on its chances for success. They’ve also used the approach to sort out technical or customer service glitches.

More casual users who don’t want to sink money into a professional system can tap simpler versions like Tweetfeel, Twendz, and Twitrratr for topic-based opinions.

—Guest-blogger Susan Karlin

If you don’t have  time to watch the video you can read recaps and quotes from the panel here, here, here, here and here.

Big thanks to Jennifer at SEE, to all of our panelists, and to the Bad Astronomer, who found time to moderate our panel while he wasn’t partying with Hollywood starlets (Phil – we kid because we love).