2012-11-12
2012-04-20
2012-01-27
2011-11-29
2011-10-11
2011-01-28
<!--break-->
Supposing a technological Singularity or something like it does occur later in this century – what’s likely to be the main technology pushing it forward?
Like Ray Kurzweil and many others, I believe the answer is: Artificial General Intelligence. Other technologies will surely play large roles, but what will really push us over the threshold and radically transform our world, will be the emergence of engineered minds with general intelligence significantly greater than our own.
Nobody fully understands AGI yet – but there’s an active community of researchers hammering away at the problem every day. And while this community shares a passion for AGI and a belief in the tractability of the problem, it’s also characterized by a wild diversity of perspectives on nearly every relevant technical and conceptual issue.
As an AGI researcher I have my own strong opinions about the best path to creating powerful AGI, but nevertheless, I always find it interesting to probe the different views of other AGI researchers. This is why, back in 2009, I collaborated with Seth Baum and my father in creating a survey of expert opinion on the timeline to AGI. And it’s why this year I’ve decided to conduct a series of in-depth interviews with a handful of AGI researchers, poking more deeply into some issues concerning the field.
This first interview in the series is with Dr. Pei Wang, who is one of the other AGI researchers I’ve worked with most closely.
After first getting his Computer Science education in China, Pei came to the US to complete his PhD with famed writer and Cognitive Scientist (he doesn’t consider himself as an AI research anymore) Douglas Hofstadter. Since that time he’s held a variety of university and industry positions, including a stint working for me in the dot-com era (he was the Director of Research at the New York AI firm Webmind Inc., where I was CTO and co-founder). Currently he’s a computer science faculty at Temple University, and pushing ahead his approach to AGI based on his NARS system (Non-Axiomatic Reasoning System). While NARS is quite different from my own OpenCog approach to AGI, it was part of the inspiration for OpenCog’s PLN
Those with a technical bent may enjoy Pei’s 2006 book Rigid Flexibility: The Logic of Intelligence
Ben: I'll start out with a more “political” sort of question, and then move on to the technical stuff afterwards….
One of the big complaints one always hears among AGI researchers is that there’s not enough funding in the field. And indeed, if you look at it comparatively, the amount of effort currently going into the creation of powerful AGI right now is really rather low, relative to the magnitude of benefit that would be obtained from even modest success, and given the rapid recent progress in supporting areas like computer hardware, computer science algorithms and cognitive science. To what do you attribute this fact?
I note that we spend a lot of money, as a society, on other speculative science and engineering areas, such as particle physics and genomics -- so the speculative nature of AGI can't be the reason, in itself, right?
Pei: We addressed this issue in the Introduction we wrote together for the book Advances in Artificial General Intelligence
Ben: Right, the list was basically like this:
* AGI is scientifically impossible
* There’s no such thing as “general intelligence”
* General-purpose systems are not as good as special-purpose ones
* AGI is already included in current AI
* It’s too early to work on AGI
* AGI is nothing but hype; there’s no science there
* AGI research is not fruitful; it’s a waste of effort
* AGI is dangerous; even if you succeed scientifically you may just destroy the world
I guess the hardest one of these to argue against is the “too early” objection. The only way we can convincingly prove to a skeptic that it’s not too early, is to actually succeed at building an advanced AGI. There’s so much skepticism about AGI built up, due to the failures of earlier generations of AI researchers, that no amount of theoretical argumentation is going to convince the skeptics. Even though now we have a much better understanding of the problem, much better computers, much better neuroscience data, much better algorithms, and so forth.
Still, I think the situation has improved a fair bit – after all, we now have AGI conferences every year, a journal dedicated to this topic, and there are at least special sessions on human-level AI and so forth within the mainstream AI conferences. More so than 10 or even 5 years ago, you can admit you work on AGI and not get laughed at. But still, the funding isn’t really there for AGI research, the way it is for fashionable types of narrow AI.
What do you think – do you think the situation has improved since we wrote that chapter in 2006?
Pei: It has definitely improved, though not too much.
Within the fields of artificial intelligence and cognitive science, I think the major reason for the lack of effort in the direction of AGI right now is the well-known difficulty of the problem. I suppose that’s much the same as the “too early” objection that you mention. To most researchers, it seems very hard, if not impossible, to develop a model of intelligence as a whole. Since there is no basic consensus on the specific goal, theoretical foundation, and development methodology of AGI research, it is even hard to set up milestones to evaluate partial success. This is a meta-level problem not faced by the other areas you mentioned – physics and biology and so forth.
Furthermore, in those other areas there are usually few alternative research paths -- whereas in artificial intelligence and cognitive science, researchers can easily find many more manageable tasks to work on, by focusing on partial problems, and still get recognition and rewards. Actually the community has been encouraging this approach by defining “intelligence” as a loose union of various “cognitive functions”, rather than a unified phenomenon.
Ben: One approach that has been proposed for surmounting the issues facing AGI research, is to draw on our (current or future) knowledge of the brain. I’m curious to probe into your views on this a bit.
Regarding the relationship between neuroscience and AGI, a number of possibilities exist. For instance, one could propose:
A) to initially approach AGI via making detailed brain simulations, and then study these simulated human brains to learn the principles of general intelligence, and create less humanlike AGIs after that based on these principles; or
B) to thoroughly understand the principles of human intelligence via studying the brain, and then use these principles to craft AGI systems with a general but not necessarily detailed similar to the brain; or
C) to create AGI systems with only partial resemblance to the human brain/mind, based on integrating our current partial knowledge from neuroscience with knowledge from other areas like psychology, computer science and philosophy; or
D) to create AGI systems based on other disciplines without paying significant mind to neuroscience data.
I wonder, which of these four approaches do you find the most promising, and why?
Pei: My approach is roughly between the above D and C. Though I have gotten inspirations from neuroscience on many topics, I do not think to build a detailed model of neural system is the best way to study intelligence.
I said a lot about this issue in the paper “What Do You Mean by AI?” in the proceedings of the AGI-08 conference. So now I’ll just briefly repeat what I said there.
The best known object showing “intelligence” is undoubtedly the human brain, so AI must be “brain-like” in some sense. However, like any object or process, the human brain can be studied and modeled at multiple levels of description, each with its vocabulary, which specifies its granularity, scope, visible phenomena, and so on. Usually, for a given system, at a lower level descriptions say more on its internal structure, while at a higher level descriptions say more on its overall regularity and outside interaction. No level is “more scientific” or “closer to reality” than the others, so that all the other levels can be reduced into it or summarized by it. As scientific research, study on any level can produce valuable theoretical and practical results.
However, what we call “intelligence” in everyday language is more directly related to a higher level description than the level provided by neuroscience. Therefore, though neuroscientific study of the brain can gradually provide more and more details about the mechanism that supports human intelligence, it is not the most direct approach toward the study of intelligence, because its concepts often focus on human-specific details, which may be neither necessary nor possible to be realized in machine intelligence.
Many arguments supporting the neuroscientific approach toward AGI are based on the implicit assumption that the neuro-level description is the “true” or “fundamental” one, and all higher-level descriptions are its approximations. This assumption is wrong for two reasons: (1) Such a strong reductionist position has been challenged philosophically, and it is practically impossible, just like no one really wants to design an operating system as a string of binary code, even though “in principle” it is possible. (2) Even according to such a strong reductionist position, the “neural level” is not the lowest level, since it is not hard to argue for the contribution to intelligence from the non-neuron cells, or the non-cell parts or processes in the human body.
Ben: Another important conceptual question has to do with the relationship between mind and body in AGI….
In some senses, human intelligence is obviously closely tied to human embodiment -- a lot of the brain deals with perception and action, and it seems that young children spend a fair bit of their time getting better at perceiving and acting. This brings up the question of how much sense it makes to pursue AI systems that are supposed to display roughly human-like cognition but aren't connected with roughly humanlike bodies. And then, if you do believe that giving an AI a human-like body is important, you run up against the related question of just how human-like an AI body needs to be, in order to serve as an appropriate vessel for a roughly human-like mind.
(By a "roughly human-like mind" I don't mean a precisely simulated digital human, but rather a system that implements the core principles of human intelligence, using structures and processes with a reasonable conceptual correspondence to those used by the human mind/brain.)
Pei: Once again, it depends on “What do you mean by AI?” Even the above innocent-looking requirement of “a reasonable conceptual correspondence to those used by the human mind/brain” may be interpreted very differently in this context. If it means “to respond like a human to every stimulus”, as suggested by using the Turing Test to evaluate intelligence, then the system not only needs a (simulated) human body, but also a (simulated or not) human experience. However, as I argued in the AGI-08 paper, if “intelligence” is defined on a more abstract level, as a human-like experience-behavior relation, then a human-like body won’t be required. For example, an AGI system does not need to feel “hungry” in the usual sense of the word (that requires a simulated human body), but it may need to manage its own energy repository (that does not require a simulated human body). This difference will surely lead to difference in behaviors, and whether such a system is still “intelligent” depends on how the “human-like” is interpreted.
Ben: Hmmm… I understand; but this doesn't quite fully answer the question I was trying to ask.
My point was more like this: Some researchers argue that human thinking is fundamentally based on a deep and complex network of analogies and other relationships to human embodied experience. They argue that our abstract thinking is heavily guided by a sort of visuomotor imagination, for example. That our reasoning even about abstract
things like mathematics or love is based on analogies to what we see and do with our bodies. If that's the case, then an AGI without a humanlike body might not be able to engage in a humanlike pattern of thinking.
Pei: The content of human thinking depends on human embodied experience, but the mechanism of human thinking doesn't (at least not necessarily so).
If a robot has no vision, but has advanced ultrasonic sensation, then, when the system has AGI, it will develop its own concepts based on its own experience. It won't fully understand human concepts, but we cannot fully understand its, neither. Such a system can develop its "math" and other abstract notions, which may be partially overlap with ours, though not completely. According to my definition, such a system can be as "intelligent" as human, since its experience-behavior relationship is similar to ours (though not the experience, or behavior, separately). By "abstract", I mean the meta-level mechanism and processes, not the abstract part of its object-level content.
Ben: Yes, but … I'm not sure it's possible to so strictly draw a line between content and mechanism…
Pei: Of course, it is a matter of degree, but to a large extent the distinction can be made. On a technical level, this is why I prefer the "reasoning system" framework --- here "object language" and "meta language" are clearly separated.
Ben: The mechanism of human thinking is certainly independent of the specific content of the human world, but it may be dependent in various ways on the "statistics" (for lack of a better single word) of the everyday human world.
For instance, the everyday human world is full of hierarchical structures; and it's full of solid objects that interact in a way that lets them maintain their independence (very different than the world of, say, a dolphin or an intelligent gas cloud on Jupiter – I wrote an article last year speculating on the properties of intelligences adapted for fluid environments). And the brain's cognitive mechanisms may be heavily adapted to various properties like this, that characterize the everyday human world. So one line of thinking would be: If some AGI's environment lacks the same high-level properties as the everyday human world, then the cognitive mechanisms that drive human-level intelligence may not be appropriate for that AGI.
Pei: Can you imagine an intelligence, either in the AI context or the Alien Intelligence context, to have sensors and motors very different from ours? If the answer is yes, then the content of their mind will surely be different from ours. However, we still consider them as "intelligent", because they can also adapt to their environment, solving their problems, etc., and their adaptation and problem-solving mechanisms should be similar to ours (at least I haven't seen why that is not the case) --- all intelligent systems need to summarize their experience, and use the pattern observed to predict the future.
I agree with you that in different environments, the most important "patterns" may be different, which will in turn favor different mechanisms. It is possible. However, the opposite is equally possible (and also interesting to me), that is, no matter in which environment, the major mechanism for adaptation, recognition, problems-solving, etc., is basically the same, and its variations can be captured as different parameter settings. This mechanism is what "intelligence" means to me, not the concrete beliefs, concepts, and skills of the system, which depend on the concrete body and environment.
Ben: On the other hand, a virtual world like Second Life also has a sort of hierarchical structure (though not as rich or as deeply nested as that of the everyday human physical world), and also has solid objects -- so for the particular two high-level properties I mentioned above, it would potentially serve OK....
Pei: Sure, though there is a difference: the hierarchical structures in the natural world is largely the result of our mental reconstruction from our experience, so there is no "correct answer", while the current virtual worlds are not that rich (which may change in the future, of course).
Ben: Also, one could argue that some cognitive mechanisms only work with a very rich body of data to draw on, whereas with a data-poor environment they'll just give nonsense results. For instance (and I say this knowing you don't consider probability central to AGI, but it's just an example), some probabilistic methods require a large number of cases in order to meaningfully estimate distributions.... In that case, such a cognitive mechanism might not work well for an AI operating in Second Life, because of the relative poverty of the data available...
Pei: In principle, I agree with the above statement, but I don't think it mean that we cannot distinguish mechanism from content.
Ben: So I am I correct in understanding that, in your view, the same basic cognitive mechanisms are at the core of AGI no matter what the high-level properties of the environment (not just no matter what is the specific content of the environment)?
Pei: Yes, though I don't mean that they are the best solution to all practical problems. In certain environments (as you mentioned above), some less intelligent mechanisms may work better.
Ben: Hmmmm… or, alternately, do you think that the cognitive mechanisms of general intelligence are tied to some high-level properties of the environment, but that these properties are so high-level that any environment one gives an AGI system is likely to fulfill them?
Pei: I'd say that intelligence works in "most interesting environments". If the environment is constant, then the traditional computational models are better than intelligent ones; on the other extreme, if the environment is purely arbitrary, and no pattern can be recognized in meaningful time, intelligence is hopeless. However, since I'm not looking for a mechanism that is optimal in all environments, it is not an issue to me.
Ben: OK, well I don’t think we can go much deeper in that direction without totally losing our audience! So I’ll veer back toward the “politics” side of things again for a bit…. Back to the nasty business of research funding! ….
As we both know, "narrow AI" research (focusing on AI programs that solve very specific tasks and don’t do anything else, lacking the broad power to generalize) gets a lot more attention and funding than AGI these days. And in a sense this is understandable, since some narrow AI applications are delivering current value to many people (e.g. Internet search, financial and military applications, etc.). Some people believe that AGI will eventually be achieved via incremental improvement of narrow-AI technology -- i.e. that narrow AI can gradually become better and better by becoming broader and broader, until eventually it's AGI. What are your views on this?
Pei: “Narrow AI” and “AGI” are different problems, to a large extent, because they have different goals, methods, evaluation criteria, application, etc., even though they are related here or there. I don’t think AGI can be achieved by integrating the existing AI results, though these tools will surely be useful for AGI.
Ben: Yeah, as you know, I agree with you on that…. Narrow AI has yielded some results and even some software that can be used in building AGI applications, but the core of an AGI system has got to be explicitly AGI-focused … AGI can’t be cobbled together from narrow AI applications. I suppose this relates to the overall the need to ground AGI work in a well-thought-out philosophy of intelligence.
Pei: Yes. One problem slowing down progress toward AGI, I think, has been a neglect among AI researchers of a related discipline: philosophy. Most major mistakes in the AGI field come from improper philosophical assumptions, which are often implicitly held. Though there is a huge literature on the philosophical problems in AI, most of the discussions there fail to touch the most significant issues in the area.
Ben: So let’s dig a little into the philosophy of mind and AI, then…, There’s one particular technical and conceptual point in AGI design I’ve been wrestling with lately in my own work, so it will be good to get your feedback.
Human minds deal with perceptual data and they also deal with abstract concepts. These two sorts of entities seem very different, on the surface -- perceptual data tends to be effectively modeled in terms of large sets of floating-point vectors with intrinsic geometric structure; whereas abstract concepts tend to be well modeled in symbolic terms, e.g. as semantic networks or uncertain logic formulas or sets of various sorts, etc. So, my question is, how do you think the bridging between perceptual and conceptual knowledge works in the human mind/brain, and how do you foresee making it work in an AGI system? Note that the bridging must go both ways - not only must percepts be used to seed concepts, but concepts must also be used to seed percepts, to support capabilities like visual imagination.
Pei: I think I look at it a little differently than you do. To me, the difference between perceptual and conceptual knowledge is only “on surface”, and the “vectors vs. symbols” distinction merely shows the choices made by the previous researchers. I believe we can find unified principles and mechanisms at both the perceptual level and the conceptual level, as a continuous and coherent “conceptual hierarchy”. Here “conceptual” means “can be recognized, recalled, and manipulated as a unit within the system”, so in this broad sense, various types of percepts and actions can all be taken as concepts. Similarly, perceptual and conceptual knowledge can be uniformly represented as specialization-generalization relations among concepts, so as to treat perception and cognition both as processes in which one concept is “used as” another in certain way.
According to this view the distinction between perceptual and conceptual knowledge still exists, but only because in the conceptual hierarchy certain concepts are closer to the sensors and actuators of the system, while some others are closer to the words in communication languages of the system. Their difference is relative, not absolute. They do not need to be handled by separate mechanisms (even with a bridge in between), but by a uniform mechanism (though with variants in details when it is applied to different parts of the system).
Ben: Certainly that’s an elegant perspective, and will be great if it works out. As you know my approach is more heterogeneous – in OpenCog we use different mechanisms for perception and cognition, and then interface them together in a certain way.
To simplify a bit, it feels to me like you begin with cognition and then handle perception using mechanisms mainly developed for cognition. Whereas, say, Itamar Arel or Jeff Hawkins, in their AGI designs, begin with perception and then handle cognition using mechanisms mainly developed for perception. On the other hand, those of us with fundamentally integrative designs, like my OpenCog group or Stan Franklin’s LIDA approach, start with different structures and algorithms for perception and cognition and then figure out how to make them work together. I tend to be open-minded and think any of these approaches could potentially be made to work, even though my preference currently is for the integrative approach.
So anyway, speaking of your own approach -- how is your own AGI work on NARS going these days? What are the main obstacles you currently face? Do you have the sense that, if you continue with your present research at the present pace, your work will lead to human-level AGI within, say, a 10 or 20 year timeframe?
Pei: My project NARS has been going on according to my plan, though the progress is slower than I hoped, mainly due to the limit of resources.
What I’m working on right now is: real-time temporal inference, emotion and feeling, self-monitoring and self-control.
If it continues at the current pace, the project, as currently planned, can be finished within 10 years, though whether the result will have “human-level AGI” depends on what that phrase means --- to me, it will have.
Ben: Heh…. Your tone of certainty surprises me a little. Do you really feel like you know for sure that it will have human-level general intelligence, rather than needing to determine this via experiment? Is this certainty because you are certain your theory of general intelligence is correct and sufficient for creating AGI, so that any AGI system created according to this theory will surely have human-level AGI?
Pei: According to my theory, there is no absolute certainty on anything, including my own theory!
What I mean is: according to my definition of "intelligence", I currently see no major remaining conceptual problem. Of course we still need experiments to resolve the relatively minor (though still quite complicated) remaining issues.
Ben: Indeed, I understand, and I feel the same way about my own theory and approach! So now I guess we should stop talking and get back to building our thinking machines. Thanks for taking the time to dialogue, I think the result has been quite interesting.
2011-01-27
<!--break-->
I was first exposed to transhumanist hopes while parading half-naked past Singularity Point tent at Burning Man 2000. Tossing back my chestnut curls, I sauntered my Adonis-like body into the tent and listened rapt as I learned that some scientists were convinced that the Seven Causes of Aging were not eternal principles but technical problems had been identified and were tractable. Leaping spryly to my feet, I cried out in a sprightly voice that through the rigorous application of scientific techniques, I would achieve "negligible senescence."
Today it's 2011. My pecs are now technically boobs. My six-pack has become a keg. Every time I lose a strand of hair from my head, another one sprouts on my ass. When I can choose between sex and a nap, I actually weigh options. After a decade of whole grains, sardines, NordicTrack and weightlifting, the ripe fuzzy apricot of my ass has taken on the texture of my scrotum.
When I get together with my fellow curmudgeons, I complain about kids today. In my day, advertisers estimated that we could watch a Devo video for an average of 2.3 seconds without needing an edit cut before we got bored and hit the remote control. Now these brats are lucky if they can get through a second-and-a-half. Where's the resonance? As a young man I aspired to cultivate the virtue of not caring what people think. Today I have perfected the vice of not giving a flying fart what people think. When I look in the mirror, I see my dad.
I've aged! Aren't we transhumanists supposed to be immune from that?
Hey, futurists. What if the only thing we can predict about the future is that future generations will think our beliefs are batshit?
We transhumanists are a diverse bunch: extropians, transgenderists, singulitarians, technogaians, abolitionists. We hold only one thing in common. As kids we read an assload of science fiction. I write fiction. Fiction is a fancy word for bullshit. If you want to believe your own bullshit, stay away from cognitive science.
You're Dumber than a Rat
Jump into a T-shaped maze with a rat, and let's see who comes out the brightest. Researchers placed a rat in a maze in the shape of a T, then alternately placed rewards on either the left or the right of the T. The rat, entering at the base of the T, was allowed to run the maze repeatedly, but it didn't know whether the reward would appear on the left or the right, so it had to guess which way to scurry. Yet the researchers rigged the food dispenser so the rewards appeared according to a secret pattern.
Researchers gave Yale undergraduates the same test, except it was a virtual maze, and instead of edible treats, the Yale students were offered brownie points.
The race was on. Which team would figure out the underlying pattern and get the most cheese? The rat realized the treats appeared more often on the left and started going left every time. The Yale students studied the data and generated hypotheses, rapidly working out deeper algorithms to the reward placement, varying their strategies as new data enriched their hypotheses.
Final tally: The rat, once it discovered a pattern, scored correctly 60% of the time. Yale undergrads, once they discovered a pattern, scored correctly 52% of the time.
What was the secret pattern? The researchers placed food on the left 60% of the time, and the right 40% of the time. That's it.
The rat swiftly zeroed in on the optimum course of action and stuck with it, scoring as well as an infinite intelligence possibly could have achieved.
The Yale students barely scored above what flipping a coin would have scored had the treat placement been truly random. Why? Because they invented patterns that weren't there. The Yale students' smartness had made them stupid.
As Jonah Lehrer says in his mindfuck primer How We Decide: "The rat didn't strive for perfection. It didn't search for Unified Theory of the T-shaped Maze. It just accepted the inherent uncertainty of the reward, and learned to settle for the option that usually gave the best outcome."
If you can't figure out the underlying pattern of rewards in a T-shaped maze, how can you figure out the pattern in technology trends? Maybe we should stop speculating about when artificial intelligence will surpass human intelligence, and start speculating about when human intelligence will surpass rodent intelligence.
But we're smarter than regular people.
Granted, more transhumanists are from MIT than Yale. Just because Yale undergraduates can't find the pattern between two possibilities doesn't mean MIT graduates can't find multiple patterns among trillions of technological, economic, political & social possibilities and accurately predict events in 2029.
For instance, a famous MIT graduate predicts that greater-than-human artificial intelligence will quickly create yet greater intelligences, causing a runaway intelligence of such benevolence that it will allow us to upload our minds into an infinitely more advanced intelligence that will suffuse the universe and turn all matter into one all-knowing self-regarding mind.
Nice! I'd like to contribute my own predictions, drawn from a close study of evolution:
Pigs evolved from worms beneath the mud. Then they evolved pig-feet, lifting their bellies a few inches from the slop. According to exponential trends, pigs will fly by 2019.
Humans evolved from monkeys. Studies of fossilized stool samples tell us hominids ate meat, including monkeys. According to exponential trends, monkeys will fly out my butt by 2029.
Do my inferences and extrapolations seem unwarranted to you? Granted, I couldn't get accepted at MIT or Yale. Unlike many transhumanists, I did not pass the IQ test that would allow me membership in Mensa. When I demanded a recount and reconsideration of my creative answers, I was informed that I qualified to be a member of Densa, which it turns out is a super-secret club available only to a select few.
Doubtless many H+ readers went to both MIT and Yale and blather at Mensa meetings. Maybe you're ten times as smart as I am. The question is, what would an intelligence ten times as smart as you think about your theories? Maybe it would be smart enough to not make predictions.
New rule, Nostradamus! No more predictions until you can beat a rat in the T-shaped maze test.
Your iPod is Smarter than you
While extending my lifespan on the Nordic Track, I set my iPod to shuffle, which is supposed to select songs randomly, yet I am maddened by the obvious patterns in the selections. The first song I hear today is often the first song I heard yesterday. The songs I listen to more often are the songs that pop up most often. "This can't be a coincidence!" I shout at random people in the gym. "Why does Steve Jobs sneak secret patterns into my iPod selection?"
After generating several conspiracy theories, I was humbled to read Jonah Lehrer's book, which informed me I am part of a vast community of investigators who have detected the same secret patterns. Now I am an apostate, because it turns out my original iPod shuffle really was truly and perfectly random. But so many customers complained that the song shuffle followed secret patterns, Apple realized the problem was chronic. Apple had to study the cognitive biases we all share in order to create a new non-random shuffle that seems random to the human brain.
Hey, transhumanists. Fifty songs randomly shuffled is not that friggin' complex. Billions of people behaving, thinking new ideas, and making stuff is pretty damn complex. Where do we get off theorizing about patterns we see in human innovation when we make up patterns on our iPods?
Your Mind is Less a Tool for Knowing than a Confabulation Machine
Don't take this the wrong way, but you're human. Your brain is exquisitely designed to make you think you have a clue. Whether you're pulling a lever on a slot machine trying to discern a pattern in those little spinning fruits, or a professional financial advisor watching the stock market convinced you're smarter than the blind index funds that consistently outperform your colleagues, the truly successful memes in our debates are those that take advantage of our belief we are discerning patterns in complexity.
Dutch psychology professor Ap Dijksterhuis says that any problem with more than four variables overwhelms the rational brain. How many variables are involved in predicting the state of civilization in twenty years? If a little bit of knowledge is a dangerous thing, what if all the knowledge your brain could possibly acquire about is a little bit of knowledge? Maybe all your learning and concluding is a dangerous thing, as most conviction throughout history has been.
I'd challenge you to look at a fluffy cumulus cloud and not see duckies and horsies and the face of your ex-boyfriend. Which is more complex? A cumulus cloud or civilization? How likely is your pattern recognition software to spot all dynamics that could possibly affect your future, discern how all those dynamics interact, and hit upon a predictive trend?
There are more things in science and technology than are dreamt of in your philosophy, futurist.
Right now, I'm at the National Air and Space Museum in Washington DC staring at a photo of the surface of Jupiter. In the swirling clouds I see a cheering megachuch, my first name in script, a row of rippling abs, a hydra-like creature that can only be described as the cockoctopus, and a cyclonic storm that looks like Uranus. Not sure what this says about my secret inclinations, but I suspect it reveals less about reality than my own mind.
Now let's look at a newspaper. See any patterns? Is the world heading toward a glorious future or hell in a hand basket? How the hell would you know? The details of the world are infinite. Your ability to process multiple streams of data is finite. Whether you're seeing a face on the moon, gods in the weather, or immortality in information technologies, this world is a Rorschach test, a mirror revealing your primal anxieties and hopes.
Listen to a man talk about politics. He is describing himself. We've all got Glenn Beck charts with circles and arrows intersecting in the private padded room of our skulls, where we compose rants we call Arguing with Idiots. Yes, you think the reason people think you are an idiot is because they are idiots.
Seeing the World is Black and White is Too Complicated for you
We are all afflicted by what Denise Minger calls the "Where's Waldo? Syndrome." Minger is an epidemiologist, sprightly writer, and cheerful eviscerator of sacred cows. She says theorists approach data with a theory that the whole story is about Waldo. You can't find Waldo unless you block out most distracting evidence of stuff you pre-assume is irrelevant. Thus you form what Denise "Mythbitchslapper" Minger calls a "Waldocentric" view of the world.
Reality doesn't need to be near as complicated as Where's Waldo? to fool your dumb ass. Google "Techdirt gorilla basketball," watch the video, and count how many times the people with white shirts pass the basketball. Merely by focusing on white shirts passing a ball, your brain relegates everything black to the background. Oh, you're skilled at counting the ball passes when you're motivated to. The only problem is you miss the big black gorilla.
What do you think? Are the dichotomies in reality any more complicated than black and white? To focus is to eliminate. The instant you search, you block out what your search does not expect to find. Take it for granted that every opinion you hold is missing some big black gorilla.
You can't even handle a simple black/white dichotomy with regard to your perception of yourself. Black men and white men were tested on a simple golf putt. When told the test measures natural intelligence, black men performed worse and white men performed better. When told the test measures natural athletic ability, black men performed better and white men performed worse. Simply by activating a racial stereotype before the test, you can change performance.
Now how much do you trust your powers of introspection? What are you good at? What are you bad at? Where did your last thought come from? How the hell would you know given the myriad blinkers activated in your unconscious that set up the parameters for how you will perceive and perform? There is no perception without pre-ception.
Most Humans have been Wrong about Virtually every Belief they've had about the World.
Most people throughout history have found meaning by agreeing to be spectacularly wrong about the universe. Cargo cults thought canned goods and tents were delivered magically from the sky. Christians think dying is actually living forever. Communists think the best way to incentivize people to work is not competition but solidarity. Libertarians scoff at the notion that some paternalistic government will make sure everything turns out for the best. They believe that a deity called the "the invisible hand" will make sure everything works out for the best. Transhumanists think science will allow us to live forever in sustained bliss.
Of course, humanity's consistent history of being delusionally wrong might end with you. The set of experts you've chosen to believe through your critical thinking skills might be the first group in history to be right about the future. After thousands of generations of disastrous declarations about human destiny, maybe somebody has finally digested enough data to make an accurate prediction. Maybe that soothsayer is you!
But then again, maybe you should join me in laughing at yourself. Goodness knows I am. To paraphrase Richard Dawkins: Most of us are skeptics about most zany beliefs people have believed. But the court jester knows all beliefs are funny.
So how do we Approach Wisdom? Let's look to the Ancients.
Socrates is a fictional character in Plato's dialogues based on a real guy he probably used to boff. Plato put words into Socrates mouth which I will translate directly from the ancient Greek, "The least clueless man is he who knows he don't know jack."
But how does Socrates know that? If he's wise, then he doesn't know anything, so where does he get off making that claim? Now I feel like that robot on Star Trek with smoke coming out of his ears when Kirk says, "I am lying."
For those of you playing the home game, wave your robot arms around and say, "But if you are lying, then you are telling the truth, which means you are lying, but if you know nothing, than how can you say that you know nothing, which means you must know something, which means your statement isn't true, which… "
The declaration works much better as a Zen koan than a philosophical aphorism, yet Plato wasn't even bright enough to point out the paradox.
Am I being unfair? Consider this Socratic statement: "I am wiser than this man, for neither of us appears to know anything great and good; but he fancies he knows something, although he knows nothing; whereas I, as I do not know anything, so I do not fancy I do."
Socrates wasn't so much a wise man as a wise ass, and his fictional dictum has survived only by people repeating it to make a point they think they know is true. But faced with such a nonsensical meme, "The only true wisdom consists in knowing that you know nothing," we can only respond like Ted from Bill & Ted's Excellent Adventure: "That's us, dude!"
If wisdom is found in the ancients, people must be getting stupider. When you consult the wisdom of the ancient philosophers, you're implying that people with less knowledge than you know more than you do. Are the ancient Greeks smarter than the modern geeks?
The fact that such an aphorism has resonated for millions of people over thousands of years demonstrates something about the human condition. You can't unravel such a self-contradictory statement by logical steps. Conclusions require thinking from the gut, a sense of joy in surprising absurdity.
Laughter is a spontaneous reaction about as controllable as a sneeze. Non-philosophical species do not laugh. Apparently brains bloated with bullshit generators require this instinct. That would explain why so many professional philosophers write with such humorlessness. No laugh kicks them out of their earnest belief that the best tool for understanding the mind is the mind. Using your mind to know your mind is like trying to taste your tongue. Studying thought by thinking is like trying to fuck yourself.
Which is where we're philosophically at, my fellow confabulists. Once you wrap your tiny mind around how utterly clueless you are, about everything… how careeningly out-of-control your decisions and beliefs are, you can't help but be plagued by the blessing of doubt, key to compassion toward those whose worldviews are so odious to your own.
Repeat this mantra: "I am stupid and I need meaning." It's what binds me to you in our argument. I find confusion intolerable; I can't control my curiosity and every human before me has been demonstrably wrong about most beliefs. Before you cite your favorite physicist or spiritualist, remember Newton calculated that Armageddon would occur in 2060, Jesus recommended self-castration (Matthew 19:12) and Gandhi drank a cup of his own pee every morning. The brief glimmers of insight that have accumulated into the majestic mess we call civilization have flashed like flecks of light on a dark sea of delusion.
Look at love, for instance.
Were you in love yesterday, and dumped today? Note how easy it is to search through all your memories and find proof that your beloved was always a shitheel. How many divorced persons do you know who describe their spouse of ten years as "evil"? Why did it take them ten years of love and commitment to figure that out? If you can demonize the person to whom you pledged your deepest self, you can demonize any stranger who threatens your worldview.
How about you, shit-for-brains? What do you wish to be true? I promise you, enough information is available for you to prove it.
Maybe science will solve the ancient riddle of death.
Then again …
Maybe death is real, and we all have to watch it coming for us, as does everybody we love, and it's intolerable, so we make up a bunch of bullshit. You, me, our allies, our enemies — we're all in this together, each equally out of his gourd with conviction.
Maybe you won't transcend being human. Maybe humanity is what we're all stuck with. No +, just H.
Maybe, just maybe, you're completely full of crap.
As evidence, observe the comments below.
Joe Quirk is the author of Exult
2011-01-26
<!--break-->
Cryonics – the use of low temperature to preserve bodies no longer maintainable by contemporary medicine, with the hope of later resuscitation – is far from a new idea. The practice goes back at least to 1967, when James Bedford was cryo-preserved by the Cryonics Society of California. But the technology has progressed tremendously since then, with organizations like Alcor and the Cryonics Institute making use of advanced techniques for preserving cryonics patients with less and less damage, increasing the odds of eventual successful resuscitation. Most exciting has been the development of vitrification, which allows the preservation of the body in a special glassy state, avoiding the damage ensuing from the traditional freezing process. And millions of dollars are being spent to keep the science and practice of cryopreservation moving forward.
If you're interested to learn more about this technology – with a view toward contributing to the science or practice, or maybe being cryo-preserved yourself – you may be interested in the Suspended Animation Conference being held later this year in South Florida. During May 20-22 2011 cryonics scientists and enthusiasts from around the world will gather at the Hyatt in Fort Lauderdale to learn about and advance the state of the art. Speakers will include cryonics pioneers such as Dr. Steve Harris, Dr. Brian Wowk, Dr. Greg Fahy, Dr. Ralph Merkle and more.
Furthermore, after the conference presentations, attendees will get the chance to visit Suspended Animation Inc.'s nearby laboratory and witness cryo-preservation technology first hand. Unlike Alcor and the Cryonics Institute, which handle both preservation and storage of patients, Suspended Animation focuses only on the first stage, getting the patient's body effectively vitrified; storage must then be handled by a separate organization (such as Alcor or CI). The concept is that this more specialized focus will allow Suspended Animation to truly excel at their task.
I have to disclose a certain bias here: I myself am signed up for cryo-preservation via Alcor, in the unhappy event that my quest for non-cryonic immortality doesn't pan out. I don't particularly want to have my body frozen – I'd rather keep breathing, or get directly uploaded into a robot or some such. But, to put it crudely, freezing seems more hopeful than rotting! If you think this sounds sensible you may wish to head to Fort Lauderdale to hear more details. And if you think it sounds crazy, hey, maybe you should check it out anyway, just in the spirit of open-mindedness. South Florida's quite pleasant in the late spring!
2011-01-25
<!--break-->
Computer scientist and futurist thinker J. Storrs Hall has been one of the leading lights of nanotech for some time now. Via developing concepts like utility fog (see the figure below) and weather machines, he has expanded our understanding of what nanotech may enable. Furthermore, together with nanotech icon Eric Drexler he pioneered the field of nano-CAD (nano-scale Computer Aided Design), during his service as founding Chief Scientist of Nanorex.
As Hall describes it, “Nanotechnology is based on the concept of tiny, self-replicating robots. The Utility Fog is a very simple extension of the idea: suppose, instead of building the object you want atom by atom, the tiny robots linked their arms together to form a solid mass in the shape of the object you wanted? Then, when you got tired of that avant-garde coffee table, the robots could simply shift around a little and you'd have an elegant Queen Anne piece instead.”
Multiple “foglets” joined together in a purpose-specific fashion to form a chair, a tool, a computer, or whatever the design spec tells them. One major role for artificial general intelligence systems in future may be to work out detailed designs for nanosystems, including utility fogs.
Hall – “Josh” to his friends – has also done a great deal to publicize nanotech concepts. He founded the sci.nanotech Usenet newsgroup and moderated it for ten years; wrote the book Nanofuture: What's Next For Nanotechnology
and led the nanotech-focused Foresight Institute from 2009-2010.
I’ve known Josh mainly via his other big research interest, artificial general intelligence (AGI). His recent book Beyond AI: Creating the Conscience of the Machine
For this interview I decided to focus largely on nanotech – I was curious to pick Josh’s brain a bit regarding the current state of nanotech and the future outlook. I found his views interesting and I think you will too! And toward the end of the interview I couldn’t help diverging into AGI a bit as well – specifically into the potential intersection of AI and nanotech, now and in the future.
While at the present time Josh’s work on AI and nanotech are fairly separate, he foresees a major convergence during the next decades. Once nanotechnology advances to the point where we have actual nanofactories, able to produce things such as flexibly configurable utility fogs, then we will need advanced AI systems to create detailed plans for the nanofactories, and to guide the utility fogs in their self-organizing rearrangements.
Ben Goertzel: Nanotechnology started out with some pretty grand visions (though, I think, realistic ones) – visions of general-purpose manufacturing and computation at the molecular scale, for example. But on your website you carefully contrast “real nanotechnology, i.e. molecular machines (as opposed to films and powders re-branded "nanotech" as a buzzword).” How would you contrast the original vision of nanotech as outlined by Richard Feynman and Eric Drexler, with the relatively flourishing field of nanotech as it exists today?
J. Storrs Hall: Feynman's vision of nanotech was top-down and mechanical. With the somewhat arbitrary definition of nanotech as used these days, progress toward a Feynman-style nanotech won't be called "nanotechnology" until it actually gets there – so you have to look at progress in additive manufacturing, ultra-precision machining, and so forth. That's actually moving fairly well these days. (See http://www.scientific.net/AMR.69-70 for a sampling of recent technical work on ultra-precision machining, or Google “ultra-precision machining” for a list of companies offering services in this area.)
Drexler's original approach to nanotech was biologically based – most people don't realize that because they associate him with his descriptions of the end products, which are of course mechanical. There's been some fairly spectacular progress in this area too, with DNA origami and synthetic biology and so forth.
I expect the two approaches to meet in the middle sometime in the 2020s.
BG: I see, and when the two approaches meet, then we will definitively have “real nanotechnology” in the sense you mean on your website. Got it. Though currently the biology approach and the ultra-precision machining approach seem quite different in their underlying particulars. It will be interesting to see the extent to which these two technological approaches really do merge together – and I agree with you that it’s likely to happen in some form.
Next question is: in terms of the present day, what practical nanotech achievements so far impress and excite you the most?
JSH: Not much in terms of stuff you can buy – although virtually every chip in your computer is "nanotech" by the definition of the NNI (the National Nanotech Initiative), so if you go by that you'd have to include the entire electronics industry.
BG: OK, not much in terms of stuff you can buy – but what progress do you think has been made, in the last decade or so, toward the construction of “real nanotechnology” like molecular assemblers and utility fog? What recent technology developments seem to have moved us closer to this capability?
JSH: Well, in the research labs you have some really exciting work going on in DNA origami, manipulation / patterning of graphene, and single-atom deposition and manipulation. In the top-down direction – "Feynman’s path" – you have actuators with sub-angstrom resolution and some pretty amazing results with additive e-beam sintering.
BG: What about utility fog? Are we any closer now to being able to create utility fog, than we were 10 years ago? What recent technology developments seem to have moved us closer to this capability?
JSH: There's actually been some research projects in what's often called "swarm robotics" at places like CMU, although one of the key elements is to design little robots simple and cheap enough to build piles of without breaking the bank. I think we're close to being able to build golf-ball-sized Foglets – meaning full-functioned ones – if anyone wants to double the national debt. You'd have to call it "Utility Hail", I suppose.
BG: OK, I see. So taking a Feynman-path perspective, you’d say that right now we’re close to having the capability to create utility hail – i.e. swarms of golf-ball sized flying robots that interact in a coordinated way. Nobody has built it yet, but that’s more a matter of cost and priorities than raw technological capability. And then it’s a matter of incremental engineering improvements to make the hail-lets smaller and smaller until they become true foglets.
Whereas the Drexler-path approach to utility fog would be more to build upwards from molecular-scale biological interactions, somehow making more easily programmable molecules that would serve as foglets – but from that path, while there’s been a lot of interesting developments, there’s been less that is directly evocative of utility fog. So far.
JSH: Right. But things are developing fast and nobody can foresee the precise direction.
BG: Now let’s turn to some concrete nanotech work you played a part in. You did some great work a few years ago with NanoRex, pioneering Computer Aided Design for nanotech. What ultimately happened with that? What's the current status of computer aided design for nanotechnology? What are the main challenges in making CAD for nanotech work really effectively?
JSH: The software we built at NanoRex – NanoEngineer-1 – is open source and can be gotten online if anyone wants to play with it.
But I think it’s really too early for nano-CAD software to come into its prime, since the ability to design is still so far ahead of the ability to build. So software like NanoEngineer-1 where you could design and simulate gadgets from Nanosystems has no serious user base. Yet. And I’d say the same is true of other nano-CAD software that has sprung up recently.
One exception is the software that allows you to design DNA origami and similar wet approaches. But most of this is research software since the techniques are changing so fast.
There will definitely be a future for advanced nano-CAD software, but any major push will have to wait for the ability to build to catch up with the ability to design.
BG: Speaking of things that may be “too early,” I wonder if you have any thoughts on femtotechnology? Is the nanoscale as small as we can go engineering-wise, or may it be possible to create yet smaller technology via putting together nuclear particles into novel forms of matter (that are stable in everyday situations, without requiring massive gravity or temperature)?
JSH: I don’t have any particular thoughts on femtotech to share. The thing about nanotech and AI are that we have natural models – molecular biology and human intelligence – that show us that the goal is possible. We don't have any such thing for femtotech.
BG: OK, fair enough. Hugo de Garis seduced me into thinking about femtotech a bit lately, and I’ve come to agree with him that it may be viable – but you’re right that we have no good examples of it, and in fact a lot of the relevant physics isn’t firmly known yet. Definitely it makes sense for far more resources to be focused on nanotech, which is almost certainly known to be possible, so it’s “just” down to engineering difficulties. Which we seem to be making good progress on!
So, switching gears yet again… you've done a lot of work on AGI as well as on nanotech – as you know my main interaction with you has been in your role as AGI researcher. How do these two threads of your work interact? Or are they two separate interests? Are there important common ideas and themes spanning your nanotech and AGI work?
JSH: There's probably some commonality at the level of a general theory of self-organizing, self-extending systems, but I'm not sure there's so much practical overlap in the near term of developing either one in the next decade. Even in robotics the apparent overlap is illusory: the kind of sensory and cognition-driven robots that are likely to be helpful in working out AGI are quite distinct from the blind pick-and-place systems that will be the first several generations of nanotech automation, I'm afraid.
BG: And what's your view on the potential of AGI to help nanotech along? Do you think AGI will be necessary in order to make advanced nanotech like utility fog or molecular assemblers?
JSH: Not to build them but to use them to anywhere near their full potential. With either utility fog or nanofactories (and also with biotech and ordinary software) you have access to a design space that totally dwarfs the ability of humans to use it. It’s easy to envision a nanofactory garage: each time you open the door you find a new car optimized for the people riding, the trip you're taking, the current price of fuel, and the weather. But who designs it; who designs the new car each time? You need an AI to do that, and one with a fairly high level of general intelligence.
BG: Yes, I agree of course. But what kind of AGI or narrow AI do you think would be most useful for helping nanotech in this way – for designing plans to be used by nanofactories? Do you think AI based closely on the human brain would be helpful, or will one require a different sort of AI specifically designed for the kinds of reasoning involved in nanotech design? If we make an AI with sensors and actuators at the nano-scale, will its cognitive architecture need to be different than the human cognitive architecture (which is specialized somewhat for macro-level sensors and actuators)? Or can the nano-design-focused AGI have basically the same cognitive architecture as a human mind?
JSH: I think the human brain, while clearly bearing the marks of its evolutionary origin, is a remarkably general architecture. And the sensors in the human body are more like what nanotech could do than what current sensor technology can do. You have a much higher bandwidth picture of the world than any current robot – and I think that's a key element of the development of the human mind.
BG: I definitely agree that the human brain gets a higher-bandwidth picture of the world than any current robot. And yet, one can imagine future robots with nano-scale sensors that get a much higher bandwidth picture of the world than humans do. I can see that many parts of the human brain architecture wouldn’t need to change to deal with nano-scale sensors and actuators – hierarchical perception still makes sense, as does the overall cognitive architecture of the human mind involving different kinds of memory and learning. But still, I wonder if making a mind that deals with quantum-scale phenomena effectively might require some fundamental changes from how the human mind works. I suppose this also depends on exactly how the nanofactories work. Maybe one could create nanofactories that could be manipulated using largely classical-physics-ish reasoning, or else one could build others that would be best operated by a mind somehow specifically adapted to perception and action in the quantum world.
But as you said about applying nano-CAD, it’s somewhat hard to explore these issues in detail until we have the capability to build more stuff at the nano scale. And fortunately that capability is coming along fairly rapidly!
So let’s talk a bit about the timing of future technology developments. In 2001 you stated that you thought the first molecular assemblers would be built between 2010 and 2020. Do you still hold to that prediction?
JSH: I'd say closer to 2020, but I wouldn't be surprised if by then there were something that could arguably be called an assembler (and is sure so to be called in the press!). On the other hand, I wouldn't be too surprised if it took another 5 or ten years beyond that, pushing it closer to 2030. We lost several years in the development of molecular nanotech due to political shenanigans [] in the early 20-aughts and are playing catch-up to any estimates from that era.
BG: In that same 2001 interview you also stated "I expect AI somewhere in the neighborhood of 2010," with the term AI referring to "truly cognizant, sentient machines." It's 2011 and it seems we're not there yet. What's your current estimate, and why do you think your prior prediction didn't eventuate?
JSH: I made that particular prediction in the context of the Turing Test and expectations for AI from the 50s and 70s. Did you notice that one of the Loebner Prize chatbots actually fooled the judge into thinking it was the human in the 2010 contest? We're really getting close to programs that, while nowhere near human-level general intelligence, are closing in on the level that Turing would have defended as "this machine can be said to think". IMHO. Besides chatbots, we have self-driving cars, humanoid walking robots, usable if not really good machine translation, some quite amazing machine learning and data mining technology, and literally thousands of narrow-AI applications. Pretty much anyone from the 50s would have said, yes, you have artificial intelligence now. In my book Beyond AI I argue that there will be at least a decade while AIs climb through the range of human intelligence. My current best guess is that that decade will be the 20s – we'll have competent robot chauffeurs and janitors before 2020, but no robot Einsteins or Shakespeares until after 2030.
BG: Yes, I see. Of course “AI” is well known to be a moving target. As the cliché says, once something can be done, it’s miraculously not considered AI anymore. We have a lot of amazing AI today already; most people don’t realize how pervasive it is across various industries, from finance to military to biomedicine etc. etc. I don’t really consider chatbots as any indicator of progress toward artificial general intelligence, but I do think the totality of narrow AI progress means something. I guess with both agree that it’s this general progress in AI-related algorithms, together with advances in hardware and cognitive science, that’s pushing us toward human-level general intelligence.
Although the question of whether robot janitors will come before robot Einsteins is an interesting one. As you’ll recall, at the AGI-09 conference we did a survey of participants on the timeline to human-level AGI – asking questions about how long it will be till we have it, but also about what kinds will come first. A lot of participants roughly agreed with your time estimate, and thought we’d have human-level AGI within the next few decades. But opinions were divided about whether the janitors or the Einsteins will come first – that is, about whether the path to human-level AGI will proceed via first achieving human-like robotic body control and then going to abstract cognition, or whether we’ll get advanced AGI cognition first, and humanlike robotics only afterwards. It seems you’re firmly in the “robotics first camp”, right?
JSH: Yes, and I explain why in my book.
I expect the progress toward true AGI to follow, to some extent, the evolutionary development of the brain, which was first and foremost a body controller whose components got copied and repurposed for general cognition. In the 70s robotics was much harder than, say, calculus, because the computers then didn't have the horsepower to handle full sensory streams but math could be squeezed into a very clean – and tiny – representation and manipulated. Nowadays we do have the horsepower to process sensory streams, and use internal representations of similar complexity. A modern GPGPU can compare two pictures in the same time an IBM 7090 took to compare two S-expressions. So robotics is getting easier almost by the day.
But the hard part of what smart humans do is finding those representations, not using them once they're programmed in. Those AI programs weren't Newtons – they didn't invent calculus, they just used it in a very idiot-savant fashion. Feynman put it this way, “But the real glory of science is that we can find a way of thinking such that the law is evident.”
We already have well-understood representations for the physical world, and we can just give these to our robots. What's more, we have a good idea of what a good janitor should do, so we can quickly see where our robot is falling short, and easily evaluate the new techniques we invent to repair its deficiencies. So I'd expect rapid progress there, and indeed that's what we see. Once we have the new, working, techniques for things like recognizing coffeemakers, we'll be able to adapt them to things like recognizing promising customers, slacking employees, and so forth that form the meat of average human intelligence.
Perhaps it will be a while later before some robot muses, without being told or even asked, that the quality of mercy is not strained, but that it droppeth as the gentle rain from heaven, upon the place beneath...
BG: Heh. Yes, I understand your perspective. As you know I don’t quite agree 100%, even though I think robotics is one among several very promising paths toward human-level AGI. But I don’t want to derail this interview into a debate on the degree of criticality of embodiment to AGI.
So let me sort of change the subject instead! What are you mostly working on these days? AGI? Robotics? Nanotech? Something else?
JSH: Back to AGI/robotics, after a detour into general futurism and running Foresight. Same basic approach as I described in my book Beyond AI and my paper at AGI-09 (see also the online video of the talk), with an associative memory-based learning variant of Society of Mind
.
Oh, and on the side, I’m trying to write a science fiction novel that would illustrate some of my machine ethics theories. And I should also mention that I have a couple of chapters in the new book Machine
Ethics coming out from Cambridge later this year.
BG: Lots of great stuff indeed! I’m particularly curious to see how the next steps of your AGI research work out. As you say, while the nanotech field is advancing step by step along both the Feynman and Drexler type paths, once we do have that nanomanufacturing capability, we’re going to need some pretty advanced AIs to run the nano-CAD software and figure out exactly what to build. Maybe I can interview you again next year and we can focus on the progress of your AGI work.
JSH: I’m looking forward to it!
2011-01-24
The nearly universal human desire to preserve youth can often motivate people to make major lifestyle changes or try the latest wonder supplement. But is it really possible to slow the rate of aging with current knowledge and technology? I argue herein that aging can be dramatically slowed by fine-tuning your longevity genes. Indeed, scientific research carried out in the last 20 years has shown that lifespan can be readily modulated by a variety of genetic or dietary strategies.
<!--break-->
In this article, I describe our efforts at Genescient LLC in Irvine, CA to develop strategies to delay aging and age-related disease. Genescient's primary business focus is on the development of pharmaceuticals for age-related diseases, but in conjunction with its spinoff firm Life Code LLC, it has provided testing services for the development of nutraceuticals based on its unique genomics platform. Our findings can be summarized as follows:
What Are the Main Effects of Aging?
Fig. 1: Aging causes an exponential increase in the annual mortality rate. The actual decline in function with age occur at the cell, organ, and systemic levels, but the impacts of this decline can differ with the individual’s genes and environment. The net result of aging in an animal population is a progressive increase in all-cause mortality and morbidity. In the case of humans, all-cause mortality is known to double every eight years after sexual maturity until it reaches an annual mortality rate plateau of about 50% over 105 years of age.
All grafted data under 110 years are from the Social Security Administration Death Master File, while data on 110 to 119 year olds are from validated human super-centenarians from the website www.grg.org.
Why Do We Age?
All life forms on earth have evolved through natural selection, which selects the best genotype for fitness in a particular ecological niche. In 1952 the British Nobel zoologist Peter Medawar proposed that aging is the simple result of the failure of natural selection to maintain fitness in older animals with declining fertility. As fertility wanes, then the chances to correct inappropriate gene expression via natural selection also decline, generating the aging phenotype. Thus, according to Medawar’s hypothesis, aging is indirectly caused by the declining forces of natural selection to select the best fitness genes for the aged animal as reproductive capacity declines. In 1957, George Williams further developed Medawar’s evolutionary theory of aging by introducing the concept of antagonistic pleiotropy, wherein a gene may promote fitness in young fertile animals (and thus be selected for) but become a liability late in life leading to a subsequent decline in fitness. Modern versions of Medawar’s and William’s evolutionary theories of aging are still widely believed today by most experts in aging science, as the theory fits well with the immense body of literature showing that natural selection is responsible for virtually all of the phenotypes present in the diverse species observed in Nature. Evolution appears to evolve a life history for each species that is best adapted to its ecological niche.
Besides its sound theoretical basis in the well-known mechanisms of natural selection, the Evolution Theory of Aging has also been directly tested in Drosophila melanogaster by Michael Rose (UCI Professor and cofounder of Genescient). If the Evolution Theory of Aging is correct, Dr. Rose predicted that he should be able to select populations of long-lived animals by simply selecting for reproductive longevity. To carry out his longevity experiment, Dr. Rose started with 5 lines of wild type Drosophila flies and selected for reproductive longevity over a 27-year period. Dr. Rose finally obtained robust Methuselah flies with a demonstrated lifespan of some 3 to 4 times that found in the non-selected control lines, while retaining fertility and sexual vitality. Genescient has carried out several independent experiments to verify that these Methuselah flies are indeed long lived compared to wild type flies. As Genescient’s VP of R&D, I carefully monitored the most recent comparative lifespan experiment done in 2010 (Fig. 2). The Methuselah flies (O populations) far outlive their unselected wild type fly B populations. The selected Methuselah O flies have some 3 or 4 times longer mean lifespan than the non-selected wild type B flies (Fig. 2). This selection experiment is a dramatic verification that evolution modulates the aging process.
Fig. 2: Breeding Drosophila for late reproduction leads to much longer lived flies. The result is as predicted by the Evolution Theory of Aging.
Studying gene expression in the wild type and Methuselah flies, Genescient has shown that several hundred genes have an altered expression in the Methuselah flies. In late 2010, Genescient sequenced the DNA of the wild type and Methuselah flies and again found that more than a hundred genes appear to be altered in the long-lived Methuselah flies.
These experimental results are fully consistent with the Evolution Theory of Aging, which predicts that aging leads to poorly functioning organisms as natural selection for optimal gene function wanes with age. In summary, we age because of the declining force of natural selection in adult life, which leads to unfit gene expression with age.
Developing Nutraceuticals That Can Extend Mean and Maximum Lifespan
If there are hundreds of genes that function poorly as we age, then one possible anti-aging strategy is to utilize wide-spectrum nutraceuticals to modify gene expression to a state consistent with greater longevity. Note that the ideal gene expression pattern is not identical to youthful gene expression, as some of the youthful gene expression is inconsistent with longevity (e.g. genes promoting rapid growth that can lead to cancer).
To develop potential wide-spectrum antiaging nutraceuticals, Genescient initially set out to identify nutraceutical compounds that would target as many of the complementary longevity pathways as possible and thereby extend Drosophila lifespan. Unfortunately, none of the single compound nutraceuticals tested appeared to significantly extend fly lifespan in our longevity screens. The typically poor longevity effects of single compounds argue against the use of drug-like therapeutics directed to a single target for longevity treatments.
At this point, I decided to test mixtures of medicinal herbal extracts, as these have had a long history of success in Chinese and Indian traditional medicine and are known to have a wide spectrum of positive effects in humans. To affect as many longevity genes as possible, I focused on complementary herbal extracts that have antioxidant, anti-inflammatory, and metabolic potential (known factors in driving aging) along with a positive effect on longevity genes and a proven history of use in traditional herbal medicine to treat a wide spectrum of diseases.
In selecting a group of herbal extracts, I did not take the traditional route of choosing an existing herbal mixture or the normal scientific route of choosing a mix of herbal extracts that target a particular disease or target. While there are many claims that a particular herbal extract is “anti-aging”, I found that these claims were too anecdotal to be believed. The screen for herbal extracts I used was novel in several ways. First, I tried to identify the best wide-spectrum herb in Chinese, Indian, or Western medicine based on its long term traditional use and data indicating that the herbal extract can target multiple longevity genes identified by Genescient or by other research groups.
In Chinese traditional medicine, Astragalus membranaceus (Huang Qi) appeared to be the best Chinese herb because of its many traditional uses and recent studies demonstrating stem cell activation and inhibition of mTOR. The mTOR inhibition has extended mouse mean lifespan by 33%. In traditional Chinese medicine astragalus is considered a true tonic that can strengthen debilitated patients and increase resistance to disease in general. Modern herbal treatments with Astragalus membranaceus root (often in concert with other herbs) are partly based on clinical trials showing benefits in strengthening immune function during viral (e.g. chronic hepatitis) or bacterial infection or in those individuals undergoing dialysis for kidney failure. Clinical trials at the US National Cancer Institute and other world centers have indicated that Astragalus can strengthen immunity and improve survival in some individuals with cancer. In western herbal medicine, Astragalus root is used to enhance immunity and to help in wound healing. Astragalus compounds have also been shown to stimulate stem cells, promote peripheral nerve regeneration in rats, and inhibit mTOR (a major longevity gene shown by extensive government studies to extend lifespan in mice).
In looking for the best herb in the Indian Ayurvedic medicinal tradition, I soon focused on the potent anti-diabetic herb, Pterocarpus marsupium. Crude extracts of Pterocarpus marsupium (Indian keno tree) bark naturally have high concentrations of pterostilbene (more than 4% by weight and extraction can get this level much higher) and have been used as a traditional herbal treatment for diabetes in India for thousands of years. More recent studies in animals show potent anti-diabetic activity. Published studies have also shown that pterostilbene is a potent anticancer compound. For example, pterostilbene, an analog of resveratrol, has dose-dependent anticancer activity in five cancer cell lines. As expected, pterostilbene is known to affect most or all of the longevity genes targeted by resveratrol, but has far greater stability and efficacy.
As an herbal medicine, Pterocarpus marsupium is popular in India for its diverse health benefits. Besides diabetes, the herb is also reported to cure a wide spectrum of ailments like skin diseases, fractures, bruises, constipation, hemorrhages, and rheumatoid arthritis. These diverse health benefits of Pterocarpus marsupium make it a clear favorite to include in a preventive herbal cocktail along with Astragalus.
Having selected two of the biggest stars in the traditional herbal medicines of China and India, I looked for an effective herb with wide-spectrum health effects from the Western herbal tradition. In this case, pine bark proanthocyanidins stand out as the best wide-spectrum herbal extracts in the Western herbal medicine tradition. Proanthocyanidins are polymer chains of flavonoids (flavan-3-ols) that were discovered by Jacques Masquelier in 1948 and have been a major therapeutic supplement in Europe since the 1980s. Most of the research and commercial success with proanthocyanidins has come from extracts of a French maritime pine bark called Pycnogenol (65 to 75% proanthocyanidins) and various grape seed extracts (80-90% proanthocyanidins).
One interesting claim of health benefits from proanthocyanidins is the hypothesis that they are responsible for the “French Paradox”, wherein the French tend to have much reduced rates of cardiovascular disease compared to other Western countries on a high-fat diet because of their high intake of red wine made with grapes. Besides their cardiovascular effects, Oligo-Proanthocyanidins (OPCs as attached units of proanthocyanidins are called) are known to have many other health benefits. For example, OPCs stabilize collagen and elastin, which are two essential proteins in connective tissues from blood vessels, muscles, and skin. OPCs are reported to reduce genetic mutations, so they have some anticancer benefits. OPCs have also been shown in clinical trials to promote blood flow and endothelial nitric oxide while reducing edema, capillary fragility, and damage from pollution, toxins, and cigarette smoke. These diverse health benefits make Pine Bark proanthocyanidins another perfect candidate to combine with wide-spectrum herbal extracts from Astragalus membranaceus and Pterocarpus marsupium bark.
To round out the above herbs, I wanted an herbal compound that provided neural protection in the brain. L-theanine (also known as gamma-glutamylethylamide, or 5-N-ethyl-glutamine) is an uncommon amino acid found preferentially in green tea. Theanine is an analog of glutamine and glutamate and can cross the blood-brain barrier to directly affect the brain. Among its psychoactive properties, theanine is reported to reduce mental stress and improved cognition and mood via its binding to the GABA brain receptors in the parasympathetic nervous system. Thus, theanine appears to increase the overall level of the brain inhibitory transmitter GABA and is reported to promote alpha wave production in the brain. Theanine also increases brain dopamine concentrations and has significant affinities for the AMPA and NMDA receptors. The NMDA receptors help control memory and synaptic plasticity. Theanine may also have positive effects on serotonin levels to promote restful sleep. In rats, theanine is neuroprotective. All of these neuroprotective properties of L-theanine make it a strong complementary addition to the three essential core herbs of the herbal mix. We named the final 4-herb mix StemCell 100, because of its positive effects on adult stem cells and have filed a patent application on this wide-spectrum nutraceutical.
Drosophila Longevity Studies Using Treatment with StemCell 100
The current StemCell 100 herbal blend has gone through extensive longevity testing with Drosophila fruit flies. The Drosophila longevity study (see Figs 3 and 4 below) included three cages of fruit flies that were treated with StemCell 100 (T1 to T3) and three cages that were untreated controls (C1 to C3). Each cage started with 500 fruit flies including 250 males and 250 females. The experiment showed that mean lifespan more than doubled with a 123% increase. That would be like the average human living to 167 years of age! While fruit flies are not people, they are more like us than you might think. Drosophila has a heart and circulatory system, and the most common cause of death is heart failure. Like humans and other mammals (e.g. mice), it is quite difficult to increase their lifespan significantly. The doubling of mean lifespan by StemCell 100 outperforms every lifespan enhancing treatment ever tested in flies – including experiments using genetic modification and dietary restriction.
Fig. 3: Mean lifespan of StemCell 100 treated and control flies
The longest living fruit fly receiving StemCell 100 lived 89 days compared to the longest living untreated control which lived 48 days. That is an increase in maximum lifespan of 85% which is the equivalent of a person living to be 191 years old! It is possible that the single longest living fruit fly lived longer for other reasons such as genetic mutation; however, there were many others that lived almost as long so it was not just an aberration. For example, the oldest 5% of the treated fruit flies lived 77% longer than the oldest 5% of the control group (see Fig. 4 below).
Fig. 4: Lifespan of last 5% survivors using StemCell 100 treated and control flies
Pilot Field Trial on Human Volunteers
A small clinical field trial with six healthy individuals was run using StemCell 100 for a period of four months (Fig. 5). The average HDL (good cholesterol) gain with treatment was 11.4 mg/dL or 25%.
Fig. 5: StemCell 100 Field Trial: Blood tests were performed before and after treatment with StemCell 100 to see if the treatment changed cholesterol or other blood chemistry profiles. Liver function and blood chemistry were the same before and after treatment in all participants, but there were small reductions in total cholesterol and LDL (bad-cholesterol) with the herbal treatment. The biggest surprise was the relatively large increases in HDL (good cholesterol) in all 6 test subjects – even the three individuals that were on statins (participants 1*, 2*, and 5*) showed large increase in HDL cholesterol.
We also checked four individuals before and after StemCell 100 treatment for changes in blood pressure (Fig. 6). There was a relatively large reduction in systolic and diastolic blood pressure. The average Systolic BP (red) dropped 12 mm of Hg with treatment, while the average Diastolic BP (blue) dropped 10 mm of Hg. High HDL cholesterol and reduced blood pressure are independent indicators of longevity, so these results suggest that the StemCell 100 may reduce all-cause mortality in humans, as is the case for Drosophila.
Fig. 6: Blood Pressure Changes: Four of the 6 above field trial participants also were checked for changes in systolic and diastolic blood pressure after treatment with StemCell 100. The observed reductions with StemCell 100 are similar to those found with anti-hypertensive proscription drugs such as the ACE inhibitors.
Conclusion
Genescient used genomic studies in Drosophila to determine that aging is modulated by over a hundred genes. We then used animal longevity assays to screen for nutrigenomic supplements that extend lifespan. We succeeded in doubling Drosophila lifespan using a novel class of wide-spectrum herbal supplements that modulate genes involved in both aging and age-related disease. The doubling of mean lifespan by StemCell 100 outperforms every lifespan enhancing treatment ever tested in Drosophila – including experiments using genetic modification and dietary restriction. With this successful demonstration of the power of Genescient’s genomic R & D system, Genescient’s proprietary genomic techniques can now be applied to developing wide-spectrum drug combinations for the age-related diseases.
To market StemCell 100, Genescient entered a joint venture with Centagen. (a co-creator of the StemCell 100 formulation) to form Life Code LLC. Genescient and Centagen have spent two years in extensive animal and human testing to optimize the herbal formulation in StemCell 100. The dosage and quality of the individual components emerged as critical factors in providing safety and efficacy. With the development complete, StemCell 100 is now available. [http://www.lifecoderx.com/ ]
While the four herbal extracts in StemCell 100 formulation have tremendous synergistic properties when properly manufactured in the optimized formulation, each of the four herbal extracts in StemCell 100 – when taken separately – have exceptional records in promoting animal and human health. For example, the individual components of StemCell 100 help support:
References
1. Yu, Q., Y.S. Bai, and J. Lin, [Effect of astragalus injection combined with mesenchymal stem cells transplantation for repairing the Spinal cord injury in rats]. Zhongguo Zhong Xi Yi Jie He Za Zhi, 2010. 30(4): p. 393-7.
2. Xu, C.J., et al., [Effect of astragalus polysaccharides on the proliferation and ultrastructure of dog bone marrow stem cells induced into osteoblasts in vitro]. Hua Xi Kou Qiang Yi Xue Za Zhi, 2007. 25(5): p. 432-6.
3. Xu, C.J., et al., [Effects of astragalus polysaccharides-chitosan/polylactic acid scaffolds and bone marrow stem cells on repairing supra-alveolar periodontal defects in dogs]. Zhong Nan Da Xue Xue Bao Yi Xue Ban, 2006. 31(4): p. 512-7.
4. Zhu, X. and B. Zhu, [Effect of Astragalus membranaceus injection on megakaryocyte hematopoiesis in anemic mice]. Hua Xi Yi Ke Da Xue Xue Bao, 2001. 32(4): p. 590-2.
5. Qiu, L.H., X.J. Xie, and B.Q. Zhang, Astragaloside IV improves homocysteine-induced acute phase endothelial dysfunction via antioxidation. Biol Pharm Bull, 2010. 33(4): p. 641-6.
6. Araghi-Niknam, M., et al., Pine bark extract reduces platelet aggregation. Integr Med, 2000. 2(2): p. 73-77.
7. Rohdewald, P., A review of the French maritime pine bark extract (Pycnogenol), a herbal medication with a diverse clinical pharmacology. Int J Clin Pharmacol Ther, 2002. 40(4): p. 158-68.
8. Koch, R., Comparative study of Venostasin and Pycnogenol in chronic venous insufficiency. Phytother Res, 2002. 16 Suppl 1: p. S1-5.
9. Rimando, A.M., et al., Pterostilbene, a new agonist for the peroxisome proliferator-activated receptor alpha-isoform, lowers plasma lipoproteins and cholesterol in hypercholesterolemic hamsters. J Agric Food Chem, 2005. 53(9): p. 3403-7.
10. Manickam, M., et al., Antihyperglycemic activity of phenolics from Pterocarpus marsupium. J Nat Prod, 1997. 60(6): p. 609-10.
11. Grover, J.K., V. Vats, and S.S. Yadav, Pterocarpus marsupium extract (Vijayasar) prevented the alteration in metabolic patterns induced in the normal rat by feeding an adequate diet containing fructose as sole carbohydrate. Diabetes Obes Metab, 2005. 7(4): p. 414-20.
12. Mao, X.Q., et al., Astragalus polysaccharide reduces hepatic endoplasmic reticulum stress and restores glucose homeostasis in a diabetic KKAy mouse model. Acta Pharmacol Sin, 2007. 28(12): p. 1947-56.
13. Schafer, A. and P. Hogger, Oligomeric procyanidins of French maritime pine bark extract (Pycnogenol) effectively inhibit alpha-glucosidase. Diabetes Res Clin Pract, 2007. 77(1): p. 41-6.
14. Kwak, C.J., et al., Antihypertensive effect of French maritime pine bark extract (Flavangenol): possible involvement of endothelial nitric oxide-dependent vasorelaxation. J Hypertens, 2009. 27(1): p. 92-101.
15. Xue, B., et al., Effect of total flavonoid fraction of Astragalus complanatus R.Brown on angiotensin II-induced portal-vein contraction in hypertensive rats. Phytomedicine, 2008.
16. Mizuno, C.S., et al., Design, synthesis, biological evaluation and docking studies of pterostilbene analogs inside PPARalpha. Bioorg Med Chem, 2008. 16(7): p. 3800-8.
17. Sato, M., et al., Dietary pine bark extract reduces atherosclerotic lesion development in male ApoE-deficient mice by lowering the serum cholesterol level. Biosci Biotechnol Biochem, 2009. 73(6): p. 1314-7.
18. Kimura, Y. and M. Sumiyoshi, French Maritime Pine Bark (Pinus maritima Lam.) Extract (Flavangenol) Prevents Chronic UVB Radiation-induced Skin Damage and Carcinogenesis in Melanin-possessing Hairless Mice. Photochem Photobiol, 2010.
19. Pavlou, P., et al., In-vivo data on the influence of tobacco smoke and UV light on murine skin. Toxicol Ind Health, 2009. 25(4-5): p. 231-9.
20. Ni, Z., Y. Mu, and O. Gulati, Treatment of melasma with Pycnogenol. Phytother Res, 2002. 16(6): p. 567-71.
21. Bito, T., et al., Pine bark extract pycnogenol downregulates IFN-gamma-induced adhesion of T cells to human keratinocytes by inhibiting inducible ICAM-1 expression. Free Radic Biol Med, 2000. 28(2): p. 219-27.
22. Rihn, B., et al., From ancient remedies to modern therapeutics: pine bark uses in skin disorders revisited. Phytother Res, 2001. 15(1): p. 76-8.
23. Saliou, C., et al., Solar ultraviolet-induced erythema in human skin and nuclear factor-kappa-B-dependent gene expression in keratinocytes are modulated by a French maritime pine bark extract. Free Radic Biol Med, 2001. 30(2): p. 154-60.
24. Van Wijk, E.P., R. Van Wijk, and S. Bosman, Using ultra-weak photon emission to determine the effect of oligomeric proanthocyanidins on oxidative stress of human skin. J Photochem Photobiol B, 2010. 98(3): p. 199-206.
25. Haskell, C.F., et al., The effects of L-theanine, caffeine and their combination on cognition and mood. Biol Psychol, 2008. 77(2): p. 113-22.
26. Owen, G.N., et al., The combined effects of L-theanine and caffeine on cognitive performance and mood. Nutr Neurosci, 2008. 11(4): p. 193-8.
27. Yamada, T., et al., Effects of theanine, a unique amino acid in tea leaves, on memory in a rat behavioral test. Biosci Biotechnol Biochem, 2008. 72(5): p. 1356-9.
28. Jia, R.Z., et al., [Neuroprotective effects of Astragulus membranaceus on hypoxia-ischemia brain damage in neonatal rat hippocampus]. Zhongguo Zhong Yao Za Zhi, 2003. 28(12): p. 1174-7.
29. Nathan, P.J., et al., The neuropharmacology of L-theanine(N-ethyl-L-glutamine): a possible neuroprotective and cognitive enhancing agent. J Herb Pharmacother, 2006. 6(2): p. 21-30.
30. Nobre, A.C., A. Rao, and G.N. Owen, L-theanine, a natural constituent in tea, and its effect on mental state. Asia Pac J Clin Nutr, 2008. 17 Suppl 1: p. 167-8.
31. Murakami, S., et al., Effects of oral supplementation with cystine and theanine on the immune function of athletes in endurance exercise: randomized, double-blind, placebo-controlled trial. Biosci Biotechnol Biochem, 2009. 73(4): p. 817-21.
32. Kawada, S., et al., Cystine and theanine supplementation restores high-intensity resistance exercise-induced attenuation of natural killer cell activity in well-trained men. J Strength Cond Res, 2010. 24(3): p. 846-51.
33. Hu, Y.C. and J.Y. Hou, [Effect of zhimu and huangqi on cardiac hypertrophy and response to stimulation in mice]. Zhongguo Zhong Yao Za Zhi, 2003. 28(4): p. 369-74.
34. Chen, K.T., et al., Reducing fatigue of athletes following oral istration of huangqi jianzhong tang. Acta Pharmacol Sin, 2002. 23(8): p. 757-61.
35. Luo, H.M., R.H. Dai, and Y. Li, [Nuclear cardiology study on effective ingredients of Astragalus membranaceus in treating heart failure]. Zhongguo Zhong Xi Yi Jie He Za Zhi, 1995. 15(12): p. 707-9.
36. Sugiura, H., et al., [Effects of exercise in the growing stage in mice and of Astragalus membranaceus on immune functions]. Nippon Eiseigaku Zasshi, 1993. 47(6): p. 1021-31.
37. Cho, W.C. and K.N. Leung, In vitro and in vivo anti-tumor effects of Astragalus membranaceus. Cancer Lett, 2007. 252(1): p. 43-54.
38. Kong, X., et al., Effects of Chinese herbal medicinal ingredients on peripheral lymphocyte proliferation and serum antibody titer after vaccination in chicken. Int Immunopharmacol, 2004. 4(7): p. 975-82.
39. Takagi, Y., et al., Combined istration of (L)-cystine and (L)-theanine enhances immune functions and protects against influenza virus infection in aged mice. J Vet Med Sci, 2010. 72(2): p. 157-65.
40. Tin, M.M., et al., Astragalus saponins induce growth inhibition and apoptosis in human colon cancer cells and tumor xenograft. Carcinogenesis, 2007. 28(6): p. 1347-55.
41. Mannal, P.W., et al., Pterostilbene inhibits pancreatic cancer in vitro. J Gastrointest Surg, 2010. 14(5): p. 873-9.
42. Paul, S., et al., Dietary intake of pterostilbene, a constituent of blueberries, inhibits the {beta}-catenin/p65 downstream signaling pathway and colon carcinogenesis in rats. Carcinogenesis, 2010.
43. Paul, S., et al., Anti-inflammatory action of pterostilbene is mediated through the p38 mitogen-activated protein kinase pathway in colon cancer cells. Cancer Prev Res (Phila Pa), 2009. 2(7): p. 650-7.
44. Suh, N., et al., Pterostilbene, an active constituent of blueberries, suppresses aberrant crypt foci formation in the azoxymethane-induced colon carcinogenesis model in rats. Clin Cancer Res, 2007. 13(1): p. 350-5.
45. Chakraborty, A., et al., In vitro evaluation of the cytotoxic, anti-proliferative and anti-oxidant properties of pterostilbene isolated from Pterocarpus marsupium. Toxicol In Vitro, 2010. 24(4): p. 1215-28.
46. Alosi, J.A., et al., Pterostilbene inhibits breast cancer in vitro through mitochondrial depolarization and induction of caspase-dependent apoptosis. J Surg Res, 2010. 161(2): p. 195-201.
