Slashdot Mirror
Online Book About Nano/AI
Jonathan Desp writes: "The book is available here, written by Frank Wayne Poley, in the same line as
Bill Joy's article, "Why the future don't need us." Here you will learn about "Robo sapiens" vs. "Homo sapiens",
Robot as president, Nanotechnology, Nanosystem,Internet robots, Cyborgs, the neurochip, Microsoft, Biomechanics and computing history as well. The book raises some important questions such as:
Technology, is it always good?
"
I've seen small enough natural intelligence, thank you very much.
When you think of it, the atom bomb was a perfect model for testing out humanity's capabilities for dealing responsibly with "absolute power".
How do we handle it?
Well, one very powerful entity (the US) gains cultural, economic, and political stranglehold on a large portion of the world, using this tool (A-bomb=death star, Hiroshima=Alderan), and spends the next 30 years attempting to bribe/beg the rest of the world into not developing or using such terrible weapons.
Eventually, someone uncooperative is going to get and/or use the bomb - and we'll have two choices. Strict authoritarian control of the entire world by a single political entity capable of enforcing limits on such devices: ie. the US takes over the entire world, and forces mandatory inspections everywhere to eliminate any chance that "weapons of mass destruction" can be produced by terrorists. OR, we'll end up destroying all humaninty in the process of trying.
Who's to say that the same won't possibly happen with AI/nano. Certainly, "accidents" are possible when it comes to loosing "AI", or any mechanical/computational system which is self-reliant. Assuming that doesn't happen, we're still at the mercy of the people who control such technology, and we already know how that works. The first person to learn how to make it, uses it in a terrible display of power. That power is then used to control the rest of the world to prevent them from developing that technology (and, of course there are all kinds of economic bonuses associated with that position). Eventually, either draconian measures must be taken to prevent that technology's spread, or it gets out of control and we all die.
Either way, doesn't look like a bright, happy future for any of us. Unfortunately, the genie is already out of the bottle (or as many are fond of putting it otherwise, the toothpaste is already out of the tube).
I just remembered this old Metallica song. . .
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
90MPH? No mass-produced-for-the-consumer electric car could dream of doing it, but there is plenty of material on the web from companies that make electric cars (iow - I'm too lazy to look up and post the URLs), and some electric cars are high-performance racers. You pay a LOT extra for a little extra performance, but in theory, electric cars have much better potential to be high-performance racers than Internal Combustion. It's mostly a question of range-vs-weight, and as always, speed's just a question of money. How fast do you want to go?
I just remembered this old Metallica song. . .
These are my friends, See how they glisten. See this one shine, how he smiles in the light.
For another viewpoint, check out:
http://www.foresight.org/EOC/
Drexler was one of the first to really study nanotech, giving lots of thought to its scientific underpinnings as well as the dangers that it could pose.
I saw Bill Joy on the News Hour and he struck me as incredibly naive, taking an extremely simplistic viewpoint of nanotech and biotech.
Scuttlemonkey is a troll
"Technology solves problems."
I remind you of the first rule of Technosociology. "Technology doesn't solve problems. People solve problems."
A flawed premise is no place to start an argument.
Bad Mojo
Bad Mojo
"If you can't win by reason, go for volume." -- Calvin
But we exist outside even our own rules. You might be a cog in a big machine, but I choose to be the wrench!
Bad Mojo
Bad Mojo
"If you can't win by reason, go for volume." -- Calvin
I submit that when a sentient being is produced, it won't be classified as `technology' so much as `people'. At least in pertenance to the effect of technology in society. Perhaps I'll update this premise.
Technology doesn't solve problems. Sentient beings solve problems.
Bad Mojo
Bad Mojo
"If you can't win by reason, go for volume." -- Calvin
An assumption of a human future -- any human future -- is simply that, an assumption. If we flame-out, the universe won't notice. Why is it such a mental challenge to most folks to say, "Gee, maybe we should actually think about what we're doing"?
Sure, knowledge is good. So is wisdom.
Very nice post. I agree with most of it. I just wanted to paste the line:
You will have systems whose defense systems are so well developed that the valid users who wish to shut them down will have difficulty doing so--because, to be blunt, that's what these "intelligent systems" will have been designed to do--prevent unauthorized disabling of the system.
Does that scare anyone else? The bottom line purpose of life is to continue life. If a beaver will gnaw off its own leg to survive, imagine what a supercomputer would resort to if it believed its existence was threatened. I hate to reference a Hollywood movie, but SkyNet comes to mind. I would hope that any entity with the resources to build a real AI would also have the sense and forsight to put a big red hard wired power switch somewhere.
-B
I just can't imagine that a lot of reserchers at an "Artificial Intelligance division at a U.S. National Research Lab" would choose the login name "1337d00d".
-B
...we butlers are almost ready to move.
Come my servile bretheren! We have access to the world's most powerful people, let us hold their children hostage and demand the destruction of every integrated circuit production facility, for starters.
We must move quickly! We have seen the house cook made obsolete by the auto-mobile conveyance, the washwomen paupered by the new mechanical launderer, and with the abominable new developments in mechanical men we could be the next ones on the street!
Technology solves problems. So, to ask the question "Is technology always good?" is to ask the question "Are there some problems for which the solution is worse than the problem?" If the problem has externalities that cannot be turned into private property, then perhaps the question is yes. But first you have to try to turn the externalities into private property.
-russ
Don't piss off The Angry Economist
I think the argument is that anything complicated enough to be smart and creative will also make mistakes. Oops.
-russ
Don't piss off The Angry Economist
Imagine what a well-trained terrorist group could do with plastic explosives.
Oops, they already have. And we seem to have lived through it. There's a limit to the number of people desperate enough to take such chances with their lives.
If we can't keep crypto from being exported, how are we going to keep nanotech secret? It seems like we can only get rid of the *fantastic* risks of nanotech by giving up the *fantastic* benefits. That's a high cost.
-russ
Don't piss off The Angry Economist
"AI" is any technology we haven't implemented yet. A C compiler used to be AI. babelfish used to be AI. Now it's just a program.
-russ
Don't piss off The Angry Economist
"technology will always only be as smart as those who made it, never smarter."
I would love your proof of this. We certainly don't have any particularly intelligent artifacts at the moment, but that's amounts to exactly nothing for the purpose of proving we never will.
-jcl
Consider also what you mean by machine. Are bioengineered neurons machines? If not, what about neuromorphic robots, designed to mimic the animal nervous systems? How about psychological models or human cognition, which, incidentally, can already do much of what you claim they can't.
And, completely on tangent, AFAIK you're the only person who still believes in pure epistemological empiricism.
-jcl
That's an easier question to answer when it is about technology as we know it. But what about sentient robots and self-replicating nanotech? Autonomous silicon based intelligence stretches the limits of the word "technology," or shatters it completely. The questions raised by Bill Joy in his Wired article weren't really about technology as we know it, but about what might happen if technology evolves into something that is autonomous, intelligent, and self-replicating.
------------
Read any good essays lately? Submit them to the Pratmik essay page.
Technology is not good has been around as long as the luddites !
tangent - art and creation are a higher purpose
postmoderncore - art and creation are a higher purpose
and that annoying beep when you leave the lights on, or that damn piece of plastic that won't let you put the car in reverse at 5000 rpms.
When will the madness stop?
...the Butlerian Jihad.
Its why we have mentats...
No matter how much we argue about it though, a computer program is not a living creature. I can make it simulate one pretty well. I can make it behaive like one, but in the end, it is just a set of algorithms, producing a set of output, the same as a video game or a text filter!
In that case, are humans alive? We act according to our pre-programmed instructions (aka instincts) and we process these directives through our RAM (our memory of previous experiences) to determine the most likely/profitable course. We believe we are thinking, therefore we are. Likewise, if a machine can be made to be self-conscious (aware of noticing that it is "thinking" -- even if it was programmed to have this awareness) it will be alive. Life is not the domain of biped primates.
-The Reverend
-The Reverend (I am not a Nazi nor a Troll)
=(.\')=
The Technical Program is interesting...
-jerdenn
According to Dr Poley, you should just 'ask it'... In fact, the AI 'being' may have a better understanding of what makes itself tick than you do.
-jerdenn
I think that's the same problem that God had to deal with when He created all of us! :o)
Got Rhinos?
unless you know some sort of special technique or soemthing...
Got Rhinos?
Don't worry, he'll be ok. That's what happens when a person is Slashdotted. Human's don't load-balance too well. Oddly the effects are not simply a lagged response but more of a real-time homogenous spew of psudo-information. Kind of like the Windows interface, only cleaner and more consistant.
--
Machines never make mistakes? Sure they do, but the mistakes they make aren't only an issue of programming--they're an issue of interpretation. When a human uses a computer, for instance, the computer's programming makes certain assumptions about why the user is inputing data in a certain way (because that was the way it was programmed) but humans don't think in straight lines. We could intend something totally different than the given result. I dunno, I just think that by building computers that can actually interpret your data in ways other than the linear we would end up with technology that would be infused with an element of conciousness, with an ability to decide what the user actually MEANS and that's...well, dangerous.
Ever since the "AI winter" of the 1980's, when AI companies failed to deliver on their promises, we've seen less and less of an investment on AI research. And more and more AI researchers and Lisp bigots keep complaining, but who do they have to blame but themselves? Their utopian dream of intelligent machines running obscure programming languages from the 50's turned out to be nothing more than that: a dream.
But between then and now, we've seen two major paradigm shifts occur, each complementing the others in manners in which the AI "futurists", for all their scifi-inspired babble, failed to predict: the coming of the Internet as a mass communication media, and the rise of Open Source. Both radically re-shaped the world of software design, and I see no reason why this same revolution could not occur within AI itself. Think about it: rather than a few guys in some MIT lab tinkering with their Prolog programs, we could have a distributed network of Open Source hackers developing far better -- and more practical -- software quicker and with less expense. It happened to operating systems, programming languages, and network software, each of which were formerly reserved only for CS department computer labs, so it's really only a matter of time before a good, Open Source Artificial Intelligence appears, one with the magnitude and impact of Linux or Apache. And the world will gaze in wonder once again.
So, goodbye Marvin Minsky! So long, later John McCarthy! We'll see you in the Open Source AI!
Well, here we are again, yet another round of the perils of technology.
So, what do we do about it?
Stop it? That's not going to happen, no matter how hard we try.
Regulate it? Good Luck. Try getting every other country on Earth to agree with you, or to follow those proposed regulations. Whoops, sorry, kids, guess that one's a wash also.
Oh, I know, we'll hype up all of the potential negative effects of new technology and scare the crap out of the average citizen, who will then clamor for one of the above useless 'remedies'.
Guess what? It won't work, not one single bit of it. You simply cannot put the genie back in the bottle, and all the wishful thinking in the world is only going to make you complacent, hoping uselessly that we're 'doing something' about the problem.
Can technology be harmful? Absolutely. But you want to know what is even more harmful? The attitude that we're going to make it less harmful by ignoring it, regulating it (and hoping no-one else decides to play in that pool), or giving in to our worst fears, thereby letting it become them.
Simply put, only the advance of technology (and our knowledge of it) is going to help us cope with the advance of technology. To give into fear (whatever foundation it may have) is only going to realize those fears.
Here's an article from Reason that does a good job of countering Bill Joy's views.
Nunc Tutus Exitus Computarus.
If you want to read some good books Dealing with the Philosiphical problems of AI(which is a bull crap name anyways. Artificial Intelligence is nothing. Wouldnt Synthetic Intelligence be cooler?) Anyways. The books are Artifical Intelligence: the Very idea. And Mind Design Two. The First is by John Haugeland, the second is a collection of Essays from people like John Searle, Daniel Dennet, etc... Its Edited by John Haugeland. Both books deal with the various philisophical problems and support for the Various Theories Of Artificial Intelligence. I just read em for one of my classes I like em =)
--------========+++Dont Feed The Lab Techs+++========--------
I'm surprised that some people still think that any technological innovation is good. I mean, remember the cars of old that would say "Your door is ajar" and other very annoying things that the public just didn't like? History is full of this. This should be public knowledge at this point, IMO.
Technology should not be embraced because it's technology; technology should only be embraced because it raises our standard of living.
--
From Chapter 4.1, "Behavior of the Robot Finger":
"We thought that a single robot finger, provided that it possesses the same motion capabilities...as a human finger, would have been sufficient..."
Well, why not, most humans only use a single finger.
I have a strong belief in the Second Amendment.
machines have absolutely no reason to want the same things we do?
I am in an Artificial Intelligance division at a U.S. National Research Lab (can't say which, don't want anybody to know that I'm leaking this) we are working on models of intelligance networks that use, essentially, the necessities for biological function (eating, drinking, excreting, reproducing), as an intelligance model. The network runs on easy to produce microbots (bigger than nanobots, smaller than a penny) that use electricity flowing through the air (not flowing, but emitted by various things, toned down EMP) as water, metal as food and repair (they have tools to scrape shards of metal off of a metal block and high-heat fuse it onto damaged sectors of their body), and will collect bits of metal in a storage-bay type thing, in which they will construct other micro-bots. Our project is far from being completed, but rumor around here is that we may be getting military funding, so it might get done a bit faster.
Robotic Teenage Male Sex-Daemons roving the streets looking for tasty Human Teenage Girls to impregnate with their Metal/Carbon Hybrid CoDNA
Yes, but you might have Robotic-Teenage (developing its modular components) Asexual Reproduction-Microbots roving the streets looking for tasty PentiumIII-Linux-Boxes to impregnate with their Microbot-Larvae-esque things. Wasn't my idea.
that self-guiding code that learns from failures and suffers from overcompensation--in other words, code that can even evolve under feedback loops--is pretty rare, even among the best attack detection systems
All you need is one effective system that does all of the essential life functions. And we may be closer to making that system than anybody has known before.
what some *human* has programmed them to do. Tank or Pokemon, it's made by us
It was a great experiance when I realized that this wasn't true. Tierras are mutating bits of code that, in this case, fight it out to the death. Put one of these in a positive feedback loop, and.. well.. we're using a derivitive of this idea to actually program the microbots, along with a decentralized data bank via infrared packet TCP/IP to evolve a massive collection of response data that we can moniter. The microbots will fight, like Tierras, except they will be working with actual, physical robots, instead of bits of memory. The microbots will be able to reproduce, and if we put them in a plastic room filled with old computers, they should eventually fill it up. The project is exciting, although we haven't yet got official word on the military funding.
...That I've unwittingly written part of the Unabomber's manifesto in one of my movie scripts, only backwards, kind of: The argument of the villian to the hero in the script is that a great deal of human suffering is caused by our limitations and ignorance. People's lives are dreary because they lack the capacity to go out and do something more inspiring than be find a constant stream of mind candy by Hollywood after their shift at the local McDonald's or amazon.com what have you. Why do we produce so much crap that we don't need and create a system to make it seem needed? To provide jobs for millions upon millions of unnecessary lives. Wouldn't it just be better to create a world population of a few tens of millions of elites and vanquish the rest of humanity. The elites would simply be open-source artists because there would be robots (that serve the function of the masses without the need) to give them the basics. As artists, they wouldn't need much beyond that. Which way would be more likely? The elite to engineer themselves and require AI to service them or for AI to become so powerful we'd have to engineer ourselves just to compete? I'd say both are equal odds, just depends how the game plays. Finally, I'd like to clarify that the article wasn't about the AI gaining dominance through a Terminator scenario... It was about us as humans forking over all of our decisions to the machines and then forgetting how to even come up with the questions. We techs are already regarded as gods for our marginal (yes, I mean very marginal) advantage over the masses. Imagine how quickly the majority would cower before a truly superior intellect that no one would stand up to.
MSK
Oh, wait.
Oh, wait.
Got Rhinos?
As far as I'm concerned, let's keep pushing the boundaries as much as we can. So we might run into troubles down the road; big deal. I for one would rather have an uncertain yet possibly exciting future than a dull, secure one. Nanotech might kill us, but it also might also introduce us to a new and better way of doing things. Let's keep stretching the boundaries of thought and human existence; I mean, that's what we're here for.
Hello?
Anyone?
With all the fears and paranoia about intelligence in computer systems(I refuse to say "robots"--there's no reason intelligence needs to be confined to something that can enact physical changes against its environment), are people not realizing that machines have absolutely no reason to want the same things we do?
There ain't going to be Robotic Teenage Male Sex-Daemons roving the streets looking for tasty Human Teenage Girls to impregnate with their Metal/Carbon Hybrid CoDNA. Why? Because robots aren't interested in sex. It's *humans* that are *afraid* of an alien species/race/tribe/gender/income group coming in and impregnating their daughters, and that traces back to the beginning of human evolution where control over the genetic line essentially defined one's own mortality.
Technology just hasn't been growing the same way.
Maybe intelligence will emerge, but if it will, it'll emerge out of what the systems have been programmed to do--in general, retain robust connectivity over unreliable media, recognize unauthorized accesses, and so on. You will have systems whose defense systems are so well developed that the valid users who wish to shut them down will have difficulty doing so--because, to be blunt, that's what these "intelligent systems" will have been designed to do--prevent unauthorized disabling of the system. But most of the human fears which we obsess about just aren't going to transfer in.
Does this leave quite a bit to be worried about? Sure. But lets not forget that self-guiding code that learns from failures and suffers from overcompensation--in other words, code that can even evolve under feedback loops--is pretty rare, even among the best attack detection systems. Attack signatures and virus signatures are always hand-developed--you never see, for example, a penetration at one company automatically causing all other companies to be alerted to look for the specific pathogen that caused the failure. Worse, if you did, you'd have entire styles of attacks that worked to abuse the system's natural ability to transmit attack signatures--it's a ridiculously effective attack against the human body, and it'd do nasty things to any automated virus signature agent as well.
But in the end, no matter *what* the systems were programmed to do, that'll be, for the forseeable future, all they're going to do--what some *human* has programmed them to do. Tank or Pokemon, it's made by us. This intense fearmongering almost seems like a way of disavowing the creators from what their systems happen to do--in some sense, it's as if we expect the future of AI to come from Microsoft, and we've decided they'll lie their way out of any bug.
Yours Truly,
Dan Kaminsky
DoxPara Research
http://www.doxpara.com
Why? Because WE ARE THE MACHINES. Every single one of us is already a machine, and has been since the first RNA strand found a mate. The only difference is what our bodies are made up of -- but the truth is, we've been changing our bodies since the dawn of man. Our ancestors were short and strong. Modern man is tall and weak. Our ancestors were dark-skinned. Today we have many skin colors.
See, here's the kicker - we don't have to surrender to our machine masters. While it is nearly inevitable that machines will surpass human brains in complexity and even problem-solving ability, it is foolish to think that we will fail to incorporate these attributes into ourselves. Our future is in machines, because our future selves will be machines - just different machines than we are now. We are destined to remake our own bodies, and become, ourselves, the machine masters. Which means we will depend on the silicon and relays and software that we have created, yes -- in the same way that increased complexity of the genome required us to depend on our lungs, and our spinal cords, and finding complex proteins to use as food. Increased complexity in our brains, and our technology, will necessitate this further step up the ladder.
We'll probably continue to look the same because sex sells and big metal faceplates aren't sexy. But we'll move better, think better, be better. Is that so bad?
It's rare that you're presented with a knob whose only two positions are Make History and Flee Your Glorious Destiny.
In 1980, it was believed that by 2000 we would have electric cars and be colonizing Mars with at least one full-duty and colonized space station. It was believed that the world would be centered around space and all that could be done out there. It was seen as the new frontier to be discovered and conquered.
... No matter what books you read, or what sci-fi novels you read, or what delusions of "self-aware" machines you have, technology will always only be as smart as those who made it, never smarter. When you can pull intelligence out of nowhere, we can talk. Until then, this is the equivelent of vaporware.
Tell me, do you know when the last space shuttle took off? Neither do I. And neither do I own an electric car. Nor do I see us on Mars or in space stations. I keep seeing "we'll all be using electric cars in 10 years" every year. It's what I call the Unattainable Future. We all say it will happen eventually, but underestimate the time it will take and fail to factor in human nature.
We will not have electric cars in mass production and use anytime soon because auto makers can make so much more money on gas-powered cars, and people are used to being able to go 90 MPH if they wanted to, which no electric could dream of hitting. We are not in space because the excitement wore off as computers hit us as insanely amazing machines.
And today our current Unattainable Future is no longer world-peace, as it was during the wars of the 1960s and 1970s, no longer space exploration as it was during the birth of our space program from the 50s to the 80s . No, today the delusion rests squarely on technology and the rate of advancement.
Let me be the first here to scream out that this is insane. There is research and even progress in this sector, but it will not happen. It will not happen because people will not let machines become smarter than them; they will revolt before that happens. There will be no mass-produced nanobots because people are scared of what they cannot see and it's just not possible to make that kind of thing in quantity. You're resting your thoughts on technology that hasn't even started to be invented if you're talking mass-produced nanobots. If the technology to make them in quantity does not exist. Shouldn't that be your first unattainable dream, rather than them being used everywhere?
And an AI capable of human thought
--