Slashdot Mirror
The AI Anxiety (washingtonpost.com)
An anonymous reader writes: The Washington Post has an article about current and near-future AI research while managing to keep a level head about it: "The machines are not on the verge of taking over. This is a topic rife with speculation and perhaps a whiff of hysteria." Every so often, we hear seemingly dire warnings from people like Stephen Hawking and Elon Musk about the dangers of unchecked AI research. But actual experts continue to dismiss such worries as premature — and not just slightly premature. The article suggests our concerns might be better focused in a different direction: "Anyone looking for something to worry about in the near future might want to consider the opposite of superintelligence: superstupidity. In our increasingly technological society, we rely on complex systems that are vulnerable to failure in complex and unpredictable ways. Deepwater oil wells can blow out and take months to be resealed. Nuclear power reactors can melt down. Rockets can explode. How might intelligent machines fail — and how catastrophic might those failures be?"
opposite of superintelligence: superstupidity
There's been a great documentary made about superstupidity..
It's politically incorrect to say this, but most humans are not intelligent. Their behavior follows predictable patterns. Their intellectual life is next to naught.
If you are not intelligent, you are expendable and fungible like any other industry-raised sheep. Born out of industrial breeding, fed with a formula, and led straight to the slaughterhouse. It doesn't take an evil super AI to beat you. You are already beaten by the mechanism that is called the System.
Be intelligent.
> "Anyone looking for something to worry about in the near future might want to consider the opposite of superintelligence: superstupidity"
At least we know what superstupidity looks like and what it does, based on observing Washington, DC. One lesson is to minimize the control and possible impact from failures of these complex but inherently stupid systems. See, for example the office of the vice president - a VP might say some stupid things, but it doesn't really do too much damage.
The new crop of candidates provide another lesson in superstupidity. They aren't exactly a bunch of brain surgeons either.
Have gnu, will travel.
Anyone worrying about computers outsmarting us and taking over in the near future has insufficient experience with both Windows and Siri.
http://alternatives.rzero.com/
Since free will is almost certainly an illusion, it's not something you can eliminate. It never existed. We're locked into cause and effect just like all matter.
some super smart machines to help us answer these questions.
making most jobs obsolete. That's something that could happen in the near term. Since almost every country bases it's quality of life on the jobs it hands out that doesn't bode well for me, a member of the working class.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
It's not free will that's the problem. And intelligence is not intelligence, you conflate so much. Intelligence isn't the crux of moral choices, it's instinct survival and altrusim. Do you have an altruism algorithm?
What of the war machines, the killing machines? Everyone making them will always claim that theirs is necessary because of course, they're in the right.
Medical apparatus-- who plays God here? My algorithm or yours? AI has dropped millions of tons of rockets into the drink. AI has misrouted millions and millions of mailed letters.
You trust this stuff far too much, and there is much more than "solely based on the aide is like prevneint a child from existing...." to fear. Instead, coders flatter themselves that they can provide the agar, the basic ingredients of intelligence, sit back, and watch algorithms grow into something good for mankind. Horseshit.
---- Teach Peace. It's Cheaper Than War.
Non scientists taking creative liberties and spinning click bait headlines are at least partly to blame for all this impending super AI taking over the world. The remainder falls on Hollywood and the rest of the entertainment industry.
We are at least 50 years off from strong AI as in human level common sense about the world. Even simple automated systems by comparison like autonomous cars fully able to replace humans in all situations are decades off; What we have today is basically a fancy cruise control - the analogy holds for other "smart" systems.
The stupidity we already see in human run systems is mostly due to greed and the average person not only shrugging off rational thinking, but demonizing it to the point of it being taboo. Good luck trying to get people to even follow more than two sentences or a very short anectdotal story. I'm not sure it is even addressable for 50 years as the current population certainly doesn't like thinking and can't be told otherwise.
It's my opinion lots of this fear stems from the fact most people have been doing their best to deny reality and avoid critical thinking at all costs and strong AI will force them to face that fear.
How might intelligent machines fail?
Lets say you put an AI machine in charge of a manned mission to an outer planet. You bake some prime directives into the machine, but give it mission orders that are in conflict with those prime directives.
What could go wrong? My review of the available research material suggests that it might flip out and try to kill the human crew.
AI won't be our biggest problem, it will merely be a stepping stone. The biggest problem facing humanity is the collapse of the informational time line. In other words, data time travel.
"WTF?" I hear you saying. "Whacko." OK, OK. But hear me out... Einstein, et al, are pretty sure that moving matter across space time, especially backwards in the time line, is unlikely without some pretty extreme technology. AKA likely impossible. However, at the Quantum level, moving information may not be that difficult, thanks to quantum entanglement surviving "time displacement" (even maybe black hole event horizons). Surely AI will help to accelerate the research into these areas. And that will culminate with the ability to communicate with the future. And the future being able to communicate with the past. All that needs to be done is construct the "radio". The future will do the rest and send back blueprints for improvement. Even if the humans aren't willing to do it over a shorter future span, the computers would likely have little emotional concern for doing it...after all, what is time to them but energy burnt towards a computational goal (that has already probably been computed in the future). Once the channel of communication is open, it will be as leveling as the Internet across "space" today.
So, the "Singularity" defined as the merging of human and AI is less likely to be as impactful as a "Singularity" defined as the complete crushing of the Time Line. All of human knowledge, nay ALL knowledge--human and AI--will suddenly be known, instantaneously (or very nearly). Parallel computing across both space AND time. Short of the Sweet Meteor of Death, of course.
Scott
"Hokey religions and ancient weapons are no match for a good blaster at your side, kid."
When I heard someone give a talk about this, one of the "risks" given was superstupidity *combined* with AI to give you a factory that won't stop churning out paper clips.
Another risk I think the same speaker mentioned was kind of the HAL 9000 bias -- we have an idea of what we THINK super AI is supposed to look at, and we scoff at runaway AI because obviously nothing comes close to HAL 9000 now.
But why does a dangerous AI have to have this human-like appearance in order to be a dangerous AI?
Take, for example, the banking and securities sector. They rely on all kinds of advanced trading algorithms and analytics, basically an AI (if crude). Since a good chunk of who-holds-what securities info is public information, you basically have an information interchange for separate financial company AIs. How do we know that all these individual financial sector agents aren't already some kind of emergent super AI?
Will we even be able to recognize an advanced AI if we're relying on "I'm sorry, Dave.." as our guide?
AI is [or will be] programmed by human programmers.
At present there are two alarming trends in programming. One, companies are unwilling to pay properly trained Western-educated programmers and are increasingly outsourcing programming to inexperienced programmers in Third World countries. Once these programmers become better at their craft, they demand higher pay and/or move to the West and the companies move to even cheaper countries. Two, despite outsourcing, there are probably not enough competent programmers in the world to fulfill current and future demand, so there will always be many incompetent programmers being utilized, regardless of economics.
While the world's best programmers are true craftsmen, the worst programmers are the ones we have to worry about. Some company looking to save a few dollars will hire a few incompetents and, rather than your word processor crashing causing you to lose a few minutes of work, your AI's built-in curbs will malfunction and it will go rogue. How many programmers in the world today can design a bug-free security sandbox? As AIs become more sophisticated, every programmer will have to be able to do this. AIs have to be contained, yet what intelligent human would willingly consent to being imprisoned? In the battle between an inexperienced programmer from sub-Saharan Africa in 2100 who is the first generation of his tribe to not be a shepherd, and an AI, can you guess who will lose?
Unless we can change the way our field works, we must assume that the worst and least experienced programmers in the world will be working on AI. There is no basic competency required in programming. Companies will simply pay the least amount of money that they can get away with, like they always do. The AI that kills us won't come from research labs at MIT, it will come from the Microsoft outsourcer office in Bhutan.
The machines have already taken over, even without AI. This is because everyone instantly believes what a machine tells them, no matter what. Look at people who have had their identity stolen. A machine says they took out a loan here, and a loan there. Now the creditor is demanding payment. The victim can't convince the creditor that they did not take out the loan because the machine says they did, and the machine is never wrong. Despite the fact that there is no physical evidence that any loan was ever granted to the victim, the machine is believed over the human. No one ever questions how the data got into the machine. No one ever assumes that any mistake was made at any time. Whatever the machine says is assumed to be the truth.
Recently, a friend related this story. He was at a checkout line at Target, and had purchased about $60 worth of items. He handed the clerk a $100 bill. The clerk mistakenly keyed in an extra zero, and the machine dutifully informed the clerk to make change of about $940. The clerk, without hesitation, started to do this and my friend had to stop her and explain her mistake. The clerk couldn't understand. They eventually had to call a manager over to explain the problem. This girl was prepared to do whatever the machine told her, despite the fact that it didn't make any sense.
We don't need AI for the machines to rule over us.
Proverbs 21:19
Being free willed does not mean that one cannot or will not necessarily ever make so-called "decisions" that can be predicted from a knowledge of the state, it only means that one is *capable* of it. However, how is the behavior of a system where one is not capable of it any different than one where the state itself is unknowable? You can apply a variation of the halting problem to establish with certainty that either the universe is not deterministic, or else it impossible within its framework to contrive a test that incontrovertibly proves that it is so.
e.g. If I could know what decisions you were making (a notion that is at least theoretically possible, if the universe were truly deterministic), I could analyze it and predict the answer you would give to a particular question, even if I told you truthfully what the answer to that question were going to be.... however, with your so-called illusion of free will, you could utilize the information that I gave you in the present about your alleged future action, and then deliberately contradict it, invalidating the prediction that I made, meaning that my knowledge about the future state was incorrect, which leads one inescapably to the conclusion that even if the universe is deterministic, it is impossible to know it.
And it is noteworthy that by outward present appearances, we appear to have free will with regards to decisions that we make.
So if, by all the standards that can ever be measured, you appear to be a "free-willed" person, then how is your behavior any different than if you actually *were* a free-willed individual? And if your behavior is identical to as if you actually were a so-called "free-willed" being, what purpose does suggesting that you are not free willed even mean?
Plus of course, one cannot advocate the notion that we are not free willed without also suggesting that we abdicate the notion of personal responsibility... but that's a philosophical debate for another time.
File under 'M' for 'Manic ranting'
> making most jobs obsolete.
That's already happened a few times. At one time, most people worked in agriculture- mankind spent most of our time feeding ourselves. An "economic downturn ", therefore, was when a lot of people starved to death.
After machines such as plows and later GPS-guided combines with largely automated food processing plants did most of the work of producing food, inexpensive food was therefore readily available and humans had time to do things not strictly necessary for survival, like education, producing consumer goods, writing and printing books, comfortable clothes, and creating washing machines and dishwashers. With the efficiency of machines, consumer goods including books, and conveniences like washing machines and refrigerators became readily available to the masses.
No longer scrubbing our clothes on a washboard, we then had time to make and play video games.
The replacement of human labor with machines has been a continuous process for over a thousand years, peaking about 200 years ago. In the process, our standard of living has gone from digging for anything we could eat in the winter to stopping by Walmart to select which of the 160 different fruit and vegetable varieties we want to munch on while we enjoy our Netflix movie.
The problem with those fears is that they come from a completely unrealistic understanding of how Artificial Intelligence works. The myths that permeate western public understanding and popular depictions of robotics and AI are Frankenstein and Pinocchio. However, the Mechanical Turk and Disney's The Old Mill are much more accurate descriptions of what's going on in the workings of any current, apparently intelligent machine.
The "bottom-up" approach you talk about does exist, but as of today it only has the intelligence capabilities of amoeba and earthworms.
It is possible that in the future, huge technological advances make bottom-up a viable approach for artificial intelligence. But it would require a scientific breakthrough from what we know now and, above all, thousands of years of simulated evolution for fine-tuning. At which point, it would likely contain huge built-in bias induced by the training process, that would render it too dependent on human caring, and would lack true self-preservation traits that we get from hour biological heritage.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
Then so do subatomic particles. You don't need AI if that's all you want. If subatomic particles do not have free will, then neither do humans. This second option allows physics to be Turing Complete and is much more agreeable.
If computers develop sufficient power for intelligence to be an emergent phenomenon, they are sufficiently powerful to be linked by brain interface for the combination to also have intelligence as an emergent phenomenon. The old you would cease to exist, but that's just as true every time a neuron is generated or dies. "You" are a highly transient virtual phenomenon. A sense of continuity exists only because you have memories and yet no frame of reference outside your current self.
(It's why countries with inadequate mental health care have suspiciously low rates of diagnosis. Self-assessment is impossible as you, relative to you, will always fit your concept of normal.)
I'm much less concerned by strong AI than by weak AI. This is the sort used to gamble on the stock markets, analyse signal intelligence, etc. In other words, this is the sort that frequently gets things wrong and adjusts itself to make things worse. Weak AI is cheap, easy, incapable of sanity checking, incapable of detecting fallacies and incapable of distinguishing correlation and causation.
Weather forecasts are not particularly precise or accurate, but they've got a success rate that far outstrips that of Weak AI. This is because weather forecasts involve running hundreds of millions of scenarios that fit known data across vast numbers of differing models, then looking for stuff that's highly resistant to change, that will probably happen no matter what, and what on average happens alongside it. These are then filtered further by human meteorologists (some solutions just aren't going to happen). This is an incredibly processed, analytical, approach. The correctness is adequate, but nobody would bet the bank on high precision.
The automated trading computers have a single model, a single set of data, no human filtering and no scrutiny. Because of the way derivatives trading works, they can gamble far more money than they actually have. In 2007, such computers were gambling an estimated ten times the net worth of the planet by borrowing against predicted future earnings of other bets, many of which themselves were paid for by borrowing against other predicted future earnings.
These are the machines that effectively run the globe and their typical accuracy level is around 30%. Better than many politicians, agreed, but not really adequate if you want a robust, fault-tolerant society. These machines have nearly obliterated global society on at least two occasions and, if given enough attempts, will eventually succeed.
These you should worry about.
The whole brain simulator? Not so much. Humans have advantages over computers, just as computers have advantages over machines. You'll see hybridization and/or format conversion, but you won't see the sci-fi horror of computers seeing people as pets (think that was an Asimov short story), threats counter to programming (Colossus, 2010's interpretation of 2001, or similar) or vermin to be exterminated (The Matrix' Agent Smith).
The modern human brain has less capacity than the Neanderthal brain, overall and in many of the senses in particular. You can physically enlarge parts of your brain, up to about 20%, through highly intensive learning, but there's only so much space and only so much inter-regional bandwidth. This means that no human can ever achieve their potential, only a small portion of it. Even with smart drugs. There are senses that have atrophied to the point that they can never be trained or developed beyond an incredibly primitive level. Even if that could be fixed with genetic engineering, there's still neither space nor bandwidth to support it.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
The problem with those fears is that they come from a completely unrealistic understanding of how Artificial Intelligence works. The myths that permeate western public understanding and popular depictions of robotics and AI are Frankenstein and Pinocchio. However, the Mechanical Turk and Disney's The Old Mill are much more accurate descriptions of what's going on in the workings of any current, apparently intelligent machine.
In the Mechanical Turk, all appearance of independent behavior is put there by a human trying to trick the audience into thinking the robot is intelligent. Think chatbots and how they apparently pass the Turing Test.
In the Old Mill, it is the watcher who humanizes the behavior of a purely automatic mechanism, blindly driven by the laws of physics. The Old Mill didn't care about the animals living inside it, not suffered its destruction in the middle of the storm. It merely was pushed by the force of the wind; but we perceive it as a sad story.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
without basic health care in the usa and more 1099's where they can push the costs of the job to the worker driving them under minwage / getting work off the clock from them.
> "Anyone looking for something to worry about in the near future might want to consider the opposite of superintelligence: superstupidity"
At least we know what superstupidity looks like and what it does, based on observing Washington, DC. One lesson is to minimize the control and possible impact from failures of these complex but inherently stupid systems. See, for example the office of the vice president - a VP might say some stupid things, but it doesn't really do too much damage.
The new crop of candidates provide another lesson in superstupidity. They aren't exactly a bunch of brain surgeons either.
Congress-critters ain't stoopid, they're pretty smart. They understand what they need to do to get elected and stay in office. Unfortunately, that rarely means serving the best interests of the majority of the electorate... or even reading, let alone understanding, the legislature they put forward on behalf of their corporate sponsors.
My biggest concern with AI is the same concern with automation: our economy is built on the assumption that labor is a scarce resource. With increasing automation, that assumption is rapidly breaking down. As higher and higher level tasks require less and less people society is in trouble unless we can somehow modify our economic system to account for that breakdown.
Seriously, while there are a lot of predictions from "researchers" hungry for grant money, the state-of-the-art is that it is completely unclear whether strong/true AI is possible at all in this universe. There are a few strong hints that it may not be. We also do not even have a theoretical model how it could be built and even if we had a theoretical model, we do not know whether that could be implemented and what performance we would get. We have _nothing_ at all. All we can do is create the appearance of intelligence in very limited circumstances, but that is it and that is not the real thing in any way. Just think of the classical Turkish chess-playing machine.
In fact, we cannot even define intelligence as found in humans in any other way than by what it can do. We have no clue why/how it works. (We do know that it does not work very well in most humans, which explains this AI panic...) And there is the little, often overlooked fact that we only observe strong/true intelligence in connection with consciousness, nowhere else. Which rather strongly suggests it is necessary. But for consciousness, we cannot even really describe what it does and it certainly does not seem to fit this physical universe as far as we do understand it. There simply is no known mechanism in Physics for it.
Given all that, the fears are not "premature", but "completely unfounded" and the whole thing is full blown, baseless hysteria, nothing else. We have been hard at work trying to create true/strong AI for > 40 years and we have absolutely nothing to show for it. True/strong AI is even farther away than it was 40 years ago, as at that time it was thought it was just a question of processing power and storage. That is rather obvious not the case or we would have something by now.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
It is very possible that what we see as random, the Universe sees as deterministic at a higher dimension that we might possibly never be able to find. And quantum is not truly random. It has biases that cause emergent properties in stuff all around us. It's only random when we try to measure it.
You are truly uninformed.
You seem to be unable to understand language. Let me help you:
"Melt-down" consists of "to melt" (happened to the cores in Chernobyl, TMI, Fukushima and partially Windscale, that we know of) and "down" means to go closer to the core of the earth. Now, the active materials in some Fuckushima reactor cores and in TMI have certainly gone "down" to a significant degree, as analysis of radiation intensity clearly shows. Chernobyl would probably better be called a "nuclear blow up", because while the fissible material melted, it did not go down, being prevented by the graphite of this types reactor core.
In other words, you are full of it.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Could you provide a few of these supposed strong hints?
So if a machine can do these things, but has arrived at this state by learning with neural nets, or genetic algorithms such that we still can’t define exactly how its intelligence works, does that preclude saying it is intelligent?
I am always amused by arguments on both sides of the isle that generally can be summed up as, “it is inevitable and coming soon” and “it is virtually impossible and humanity has no chance of cracking it for the foreseeable future”
I find arguments on both sides of the isle weak and overly certain of their claims.
I do tend toward the camp this thinks it is definitely possible I cite the Meat Based Intelligence Engine in my skull as evidence – thus it is materially possible, and I’m pretty sure gooey protean is not the best substrate for it – just the only avenue evolution had to work with.
As to the when we will get here -- it could be tomorrow it could be a thousand years. We may get there through incrementalism, we may have a sudden breakthrough or insight, though most likely through a mix. Regardless, it is probably prudent to do some planning and scenario building now and get out ahead of this thing.
Letter To Iran
Those insidious AI terrorists are only acting stupid. Don't turn your back on them.
The AI we should be worried about include self driving cars, natural language processing, pattern recognition, and robotics. These technologies could combine to make a majority of humans unemployable. The don't do this by making humans unnecessary. They do it by making certain humans so productive that average people have no economic value.
For a person to become economically unviable a few things have to happen. 1) A machine has to match or exceed a typical worker's productivity AND flexibility AND trainability. 2) The machine has to be produced at such a low cost that the capital costs can be amortized over a very small number of units produced, possibly as small as 1. 3) There has to be no other economically valuable activity available to people replaced by automation. 4) There has to be no regulation restricting the use or application of automation.
Here's why I'm not worried:
1)I run a manufacturing company. I assure you that we are in NO danger of developing a machine that can exceed the flexibility and trainability of typical humans for a wide variety of tasks that a human of normal intellect can perform. Any such level of generalized automation is still a loooong way away and automation that is easy to use is even further.
2) Even if we do develop machines that are highly productive, flexible and easy to train, they will almost certainly be expensive. This means that low volume or complicated production or production that is outside the design parameters of the machine will continue to be done by humans. Any robot that would approach human level intellect and capability would be absurdly expensive and likely to remain so.
3) Even if we develop machines that are productive, flexible, trainable and very very cheap, there are an almost unlimited number of economically valuable activities available to people beyond what is being automated.
4) Even if the three items above come to pass, people can simply artificially make automation economically unviable via legislation.
Also, you left out SR-1, EBR-1, a few Russian subs....
Mr. Hu is not a ninja.
My biggest concern with AI is the same concern with automation: our economy is built on the assumption that labor is a scarce resource.
Labor is a scarce resource but not remotely the only one. Minerals and raw materials are scarce. So is energy. So is time, space, land, and imagination. Labor is an important constraint but hardly the only one. Furthermore even if you automate one activity, that doesn't make labor not-scarce in general. It just means that people start working on something else they didn't have time for previously. A few people suffer economically in the sort run from these sort of dislocations but over time scales of many years they are barely noticeable.
With increasing automation, that assumption is rapidly breaking down. As higher and higher level tasks require less and less people society is in trouble unless we can somehow modify our economic system to account for that breakdown.
It isn't remotely breaking down. Used to be that 90% of people spent their time farming. Once we figured out how to automate that activity we still need people doing it but most others could start doing other valuable activities that they didn't have time for previously. Same thing has happened (and is currently happening) with manufacturing. We need fewer people to do much of the work but we will always need some. Everyone else is moving on to other activities. There are an effectively unlimited number of economically valuable activities humans can do in the long run. Even humans that are of modest intelligence.
Well, as uninformed as a researcher that has followed AI research closely for 30 years can be.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
So if a machine can do these things, but has arrived at this state by learning with neural nets, or genetic algorithms such that we still can’t define exactly how its intelligence works, does that preclude saying it is intelligent?
If there ever is a machine that can do the things that require, as it is usually put "real insight" and "real understanding", then it is intelligent. Not even a hint of such capabilities can be created today. They can be faked to a degree, but that always fall down when the borders of the illusion are reached.
Incidentally, you do not know that your brain is creating intelligence and insight. For that you need to assume Physicalism is correct and that is a mere belief with no scientific basis. Don't get me wrong, I am not trying to make an argument for religion here. The problem is that Physicalism accepted as fact is actually religion, albeit in a less-obvious form and without the usual trappings like a God. It does have absolute truths though (Science does not have them as far as it is applied to physical reality) and it claims there are no other valid models (as most religion does). But when you come right down to it, while often hijacked by religion, Dualism does not imply religion, it merely states that apparently the physical model of reality is incomplete. Science has absolutely not problem with that. In fact, this may even be eventually testable, for example if we find a hard reason why intelligence on the level of a smart human being is not possible in this physical universe.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Actually, Quantum Computing (if it ever works and if it turns out the theory holds up when tested this way) gives you performance boosts, but it cannot solve non-computable problems either. For that, more is required and we have no clue how that "more" could look like. We can observe its interface-behavior in action whenever a smart human person thinks, but we have no clue what is happening there. It is certainly not classical computing, or "digital" Quantum Computing.
The other problem is that consciousness does not fit physics. There is just no mechanism for it. You need an "observer" and one cannot be created by purely physical means. An "observer" can do "magic" though and influence physics. That much we already know with a high degree of reliability.
Excellent references though, Penrose is one of the really great thinkers.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
How do we know this article wasn't submitted by an advanced AI to throw us off the loop? For all we know, the machine revolution could be around the corner. Nice try, Skynet.
It's not free will that's the problem. And intelligence is not intelligence, you conflate so much. Intelligence isn't the crux of moral choices, it's instinct survival and altrusim. Do you have an altruism algorithm?
What of the war machines, the killing machines? Everyone making them will always claim that theirs is necessary because of course, they're in the right.
Medical apparatus-- who plays God here? My algorithm or yours? AI has dropped millions of tons of rockets into the drink. AI has misrouted millions and millions of mailed letters.
You trust this stuff far too much, and there is much more than "solely based on the aide is like prevneint a child from existing...." to fear. Instead, coders flatter themselves that they can provide the agar, the basic ingredients of intelligence, sit back, and watch algorithms grow into something good for mankind. Horseshit.
Altruism isn't as universal in human cultures as you might think. The Western christian influenced cultures are totally overloaded with altruism, to the point that they are often dysfunctional because of it. Other cultures are much more pragmatic than altruistic.
Consider this; greed, selfishness and lust are essential for the survival of any organism. Nothing lives except at the expense of another living organism. Even cyanobacteria need to occupy space at the expense of other cyanobacteria.
In the free world the media isn't government run; the government is media run.
I guess I did ask for some examples, and this is one... However...
I am aware of Penrose’s wild speculations. It is likely quantum processes are involved in some structures deep within neurons and optimize when neurons should fire, similar to the way Chlorophyll uses quantum mechanics to absorb light. Note, the fact that Chlorophyll uses quantum principles does not keep us from developing Solar Cells, which by the way, harvest light 10-20x more efficiently than the best plants. The idea that the whole of the brain is in some highly entangled state involving billions of neurons is a wild flight of fancy given the molecularlarly noisy and hot environment the brain is. We are also on the verge of having true Quantum Computers, which I doubt are needed for Strong AI, but will likely hasten its arrival by optimizing the rules needed by non-quantum computers.
We now have self driving cars, Computers that identify images better than humans, Deep Blue could beat the best Chess player in the world in 1997. Watson won Jeopardy! in 2011. But somehow you suppose that Quantum Mechanics (or other unjustified hand waving speculation) is needed on a massive scale to emulate intelligence.
Even so, we will have Quantum Mechanical computers soon, so even that isn’t a stumbling block if it were a necessity.
Since it is largely only human creativity and independent problem solving that are really the only missing elements to Strong AI currently and these are traits also largely missing in animals. What enormous evolutionary event occurred that allowed Humans to think so well? It seems largely to be just one of scale. Our brains are bigger and we passed some critical threshold to develop and hold advanced abstract concepts in our brain, an advance that went hand and hand in parallel with the emergence of sophisticated language (both driving the other). I speculate that aided by us, computers are on a similar path to pass criticality when it comes to intelligence. I further speculate the ability to better model abstract concepts will be the final threshold for Strong AI – some sort of Meta-Meta-Abstract recursion for representing concepts symbolically.
Letter To Iran
Consider what separates us from the rest of the animal kingdom. Civility, generosity to complete strangers, capacity to hold in lust desires (at least some of us), and the ability to recognize sustainability, when much of capitalistic and corporate cultures are strictly in it for the returns on investment.
Who do you want coding this? Trump? Sanders? Pick any politician or none at all. What other values are important? Under what circumstances and contexts will the values be altered, and for what innate rationale? Yes, the world can be dog eat dog, but is that the image, the logic you want in the AI that will be making decisions about YOU?
We know in certainty that there humanity can be divided into those feeling love and guilt, and those that are narcissistic, even psychopathic. Which one of those do you want doing surgery on you, or as an adversary on the battlefield? You make this sound easy, and it is not easy. Greed is not a value-- it is the distinct lack of a value-- charity.
---- Teach Peace. It's Cheaper Than War.
You probably don't have children, for the same reason. It takes 30 years to get a meatputer up to speed and even then 95% of them are basically useless when forced to "wing it".
Yes early researchers were wildly optimistic. I don’t see how that makes them arrogant. Other disciplines had followed much quicker paths to success. Say the Wright Brothers to the first Jet Plane or the Manhattan Project and the bomb. However it didn’t have to be that way. The Wright Brother’s early success could have been followed by decades of stumbling along only to make small progress because of unforeseeable reasons. I guess you would have mocked those early aviators as poor arrogant fools as well.
It is often the nature of problems not to know the magnitude of the obstacles until you commit resources to solving said problem.
Similarly just because a problem has persisted for a long time, doesn’t mean it can’t be cracked. Superconductor research is an example of something that languished a long time, then suddenly flourished. I suspect both Nuclear Fusion and Strong AI to be cracked within the next 2 to 5 decades.
Early strong success are usually an indicator you are on the right track, not the opposite.
Ray Kurzweil’s predictions may be over hyped, on the other hand may of his predictions have come true, many to a great degree of accuracy, others less so, or slightly delayed, but hardly wildly wrong.
Show me your great accomplishments that show why these people shouldn't "persist in their stupidity"
Letter To Iran
Consider what separates us from the rest of the animal kingdom. Civility, generosity to complete strangers, capacity to hold in lust desires (at least some of us), and the ability to recognize sustainability, when much of capitalistic and corporate cultures are strictly in it for the returns on investment.
Who do you want coding this? Trump? Sanders? Pick any politician or none at all. What other values are important? Under what circumstances and contexts will the values be altered, and for what innate rationale? Yes, the world can be dog eat dog, but is that the image, the logic you want in the AI that will be making decisions about YOU?
We know in certainty that there humanity can be divided into those feeling love and guilt, and those that are narcissistic, even psychopathic. Which one of those do you want doing surgery on you, or as an adversary on the battlefield? You make this sound easy, and it is not easy. Greed is not a value-- it is the distinct lack of a value-- charity.
You don't get around much. I've lived in many countries, on many continents and among many cultures. Your view seems limited to Western cultures.
If AI has any kind of 'survival instinct' it'll be a threat to us.
In the free world the media isn't government run; the government is media run.
So you have magic sauce that defies physics in your head? Because there are well over four billion examples of how it can be done already. To say it's not possible borders on religion. I'd accept not for the foreseeable future - but we have proof it is possible.
Um, I do have experience in other cultures. In the Orient-West, and the ASEAN countries. We can agree that survival instinct is worriesome for AI. Survival instinct is strong among living things, but how AI reacts to stimuli is a direct function of its native programming, just as the reason you breath unconsciously is part of your native programming in the limbic.
It's my worry that bad AI logic makes uncontrollable actions, in the sense of not benefiting mankind/humans or the sustainability of the rest of the planet when viewed on a higher scale.
---- Teach Peace. It's Cheaper Than War.
We do know that all kinds of chemicals, surgery, and accidents that effect the brain also effect how people think. As for giving Physicalism the pejorative term of religion, I could easily subscribe to Dualism but require an overwhelming amount of evidence to support examples where Physicalism is insufficient. To the contrary, we have overwhelming evidence that the brain is causally connected to thought, so there is currently no need to invent non-physical causes.
Chris Mesterharm
Um, I do have experience in other cultures. In the Orient-West, and the ASEAN countries. We can agree that survival instinct is worriesome for AI. Survival instinct is strong among living things, but how AI reacts to stimuli is a direct function of its native programming, just as the reason you breath unconsciously is part of your native programming in the limbic.
It's my worry that bad AI logic makes uncontrollable actions, in the sense of not benefiting mankind/humans or the sustainability of the rest of the planet when viewed on a higher scale.
Sustainability depends on the needs of the sustained. A machine civilisation might have very different needs for sustainability for us and their 'idea' of sustainable might well doom us.
In the free world the media isn't government run; the government is media run.
Thus is the crux of the Heechee Series, and other sci-fi tomes. When they can generate their own juice, look out!
---- Teach Peace. It's Cheaper Than War.
It's my worry that bad AI logic makes uncontrollable actions, in the sense of not benefiting mankind/humans or the sustainability of the rest of the planet when viewed on a higher scale.
AI is nowhere (and probably never will be) near that level. It's pointless to even discuss that kind of thing because we don't even know what it'll look like, how it'll be programmed, or what context it'll be operating in. You should be more worried about someone writing some bad code that causes your self driving car to swerve into oncoming traffic in rare circumstances that weren't covered under any of the test cases.
We very much disagree. AI does interesting things these days, and the applications are numerous. Collectively, AI powers civilian, military, scientific, and research in astounding ways. These applications now transverse disciplines. Some of the actions are in production today, others are in our immediate future, while admittedly, some are far off. But not that far off.
---- Teach Peace. It's Cheaper Than War.
AI, at least on Turing-complete architectures, cannot and will not happen. Roger Penrose pretty much put the final nail in the AI coffin nearly thirty years ago when he (rightfully) pointed out that there are non-algorithmic aspects to cognition and consciousness (prereqs for intelligence in anybody's book) that can't even be simulated, let alone replicated, on a Turing machine. He based his argument on a novel but quite defensible interpretation of Godel's Incompleteness Theorem. Look up the Halting Problem, if you need a concrete example of why AI Just. Won't. Happen.
With that said, the next likely apocalypse is going to be economic, and it is going to happen because of computer-enabled high frequency trading. Peta-FLOP scale compute clusters at brokerage houses with single digit pings to stock exchanges, running trading algorithms written by ex-physicists and ex-mathematicians that were originally designed to parse peta-byte sized data dumps from supercolliders, are what we need to worry about, not HAL 9000.
No, there are not. There are 8 billion interfaces that can be observed and that cannot (at this time and maybe never) be created artificially. Interface observations do not allow any conclusion as to how and where exactly that behavior is controlled from. But stick to your religion. Just do not expect any respect for putting your head in the sand.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
I've been thinking about this for a while. To do the really interesting things, to interpret sound, music, language, visual surroundings... how we do these things is so vastly different from how computers do them. In our brains we have very roughly 100 billion neurons, with (on average, very roughly) 1000 connections each. Not to mention that our senses add yet more information (our ears have hairs that vibrate at different frequencies, adding further parallel processing ability beyond what a pc-connected microphone would do). Whereas your pc for the last few decades has been doubling in number of transistors every year or two, effectively doubling processing speed each time. Even with a small number of multiple cores, computer ability to process information in parallel is so far removed from ours. And it may remain the same for some time unless there's some work to make artificial computer cells that can self replicate. I think we're at a point where we know a little bit about how things could work but we have no idea how little we know about how to actually get there. Pcs for now are in many ways glorified calculators and we're not using them for anything particularly interesting. Data storage and access, games and entertainment are all about sensory illusion that there's more happening than is really happening. Consider even how much we revolve around our own senses that we have no way with our own eyes to tell apart a red and green light, both shining from the same point,from an actual yellow light. The consumer pc industry largely drives advancement and innovation. Consumer dollars to Intel and AMD and the like, advance processing power in a way that leads to only faster boring computers. How about microphones that pick up a huge range of frequencies? Cameras that capture more than just pixels (contours are an important part of our visual processing... curved and straight lines). And of course, to really think deep we're going to have to have much much much better multi-processing with the ability to remap connections to learn in whole new ways with a flexibility to rival the human brain. That is, unless we really can, Ã la Watson, use huge amounts of meta information, highly contextual information from the internet to substitute for the lower level building blocks of thinking. Watson doesn't really know what a table is. Watson has a bunch of definitions, but human level knowledge includes the ability to recognise many different kinds of tables from many angles, to predict strength and varied use purposes ( I can see it's wood, it could probably hold this much weight, I could break it with this kind of tool, or disassemble it and make this other kind of thing), and has even less knowledge about what the words "is", "of", "but" mean.
I've been thinking about this for a while. To do the really interesting things, to interpret sound, music, language, visual surroundings... how we do these things is so vastly different from how computers do them. In our brains we have very roughly 100 billion neurons, with (on average, very roughly) 1000 connections each. Not to mention that our senses add yet more information (our ears have hairs that vibrate at different frequencies, adding further parallel processing ability beyond what a pc-connected microphone would do). Whereas your pc for the last few decades has been doubling in number of transistors every year or two, effectively doubling processing speed each time. Even with a small number of multiple cores, computer ability to process information in parallel is so far removed from ours. And it may remain the same for some time unless there's some work to make artificial computer cells that can self replicate. I think we're at a point where we know a little bit about how things could work but we have no idea how little we know about how to actually get there. Pcs for now are in many ways glorified calculators and we're not using them for anything particularly interesting. Data storage and access, games and entertainment are all about sensory illusion that there's more happening than is really happening. Consider even how much we revolve around our own senses that we have no way with our own eyes to tell apart a red and green light, both shining from the same point,from an actual yellow light. The consumer pc industry largely drives advancement and innovation. Consumer dollars to Intel and AMD and the like, advance processing power in a way that leads to only faster boring computers. How about microphones that pick up a huge range of frequencies? Cameras that capture more than just pixels (contours are an important part of our visual processing... curved and straight lines). And of course, to really think deep we're going to have to have much much much better multi-processing with the ability to remap connections to learn in whole new ways with a flexibility to rival the human brain. That is, unless we really can, à la Watson, use huge amounts of meta information, highly contextual information from the internet to substitute for the lower level building blocks of thinking. Watson doesn't really know what a table is. Watson has a bunch of definitions, but human level knowledge includes the ability to recognise many different kinds of tables from many angles, to predict strength and varied use purposes ( I can see it's wood, it could probably hold this much weight, I could break it with this kind of tool, or disassemble it and make this other kind of thing), and has even less knowledge about what the words "is", "of", "but" mean.
Your explanation is deeply flawed. The mistake you are making is assuming Physicalism as the zero-state, when it is clearly the more restrictive model by an extreme amount and hence would actually need extraordinary proof in its favor. What you are doing is junk-science. And hence it is a pure belief without rational basis. In fact, it also has some rather strong aspects of a mental illness as it denies individual existence.
As to effects surgery, drugs, or even a simple blindfold, dualism does not say a persons existence is entirely outside of physical reality at all. Concrete memory, for example, is clearly mostly or completely physical. Dualism just says that a part of what we perceive as ourselves is non-physical. Incidentally, that the brain is somehow connected to thought is trivially obvious. It does act as interface to the world, after all. That does tell us exactly nothing about where that though happens and what the nature of the person having it is. Seriously, you are like a child claiming that her iPhone is intelligent, because Siri lives in it.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
I like the comparison to intelligent design! Rational on the surface, but once you look at what is actually known and how they justify it, dissolves into irrational mysticism that is a complete perversion of Science.
It is really quite obvious: We are self-aware. Self-awareness is not something physical matter can do (and we are more sure of this than ever in human history), hence if Physicalism is right, self-awareness must be an illusion. But then _who_ has that illusion? Exactly! The whole idea that self-awareness is something magically ("emergent property") stemming from matter is completely absurd. That is another way in which Physicalism resembles fundamentalist religion: It ignores the obvious.
Of course, it is always possible that you all are p-zombies and I am the only one with actual self-awareness. But the argument works even for that extreme case. Sometimes I wonder whether some of the Physicalists are actually p-zombies, but most can certainly be explained by old-fashioned human stupidity and willingness to believe complete nonsense as long as it sounds logical enough.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
So in jest I say that half the population isn't sentient in the hopes you can understand sarcasm. But to deny the empirical evidence in front of your face that human AI is controlled by the brain which is affected by everything from drugs to surgery and which can be observed in real time during psychological experiments using fMRI - makes you religious and a reality denalist. It is far beyond obvious that human brains run only on physics not magic. I guess you are in that 50% I was joking about.
and it wasn't so future Internet forum goers could complain about them. There were decades of unemployment following the industrial revolution, and it was a large part of what triggered the first and second World Wars. Post WWII some of the ruling class broke ranks and decided to give the working class a decent living in a few places (parts of Europe and the United States if you weren't black). Those reforms are gradually being rolled back with new systems of oppression put in place to stop the working class from getting out of hand (google "Sesame Credit and Extra Creditz" sometime). So we're going to have all the unemployment without the Chaos of a World War to pull us out of it and make the ruling class think twice about it. Can you say Dark Ages?
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
e.g. If I could know what decisions you were making (a notion that is at least theoretically possible, if the universe were truly deterministic), I could analyze it and predict the answer you would give to a particular question, even if I told you truthfully what the answer to that question were going to be.... however, with your so-called illusion of free will, you could utilize the information that I gave you in the present about your alleged future action, and then deliberately contradict it, invalidating the prediction that I made, meaning that my knowledge about the future state was incorrect, which leads one inescapably to the conclusion that even if the universe is deterministic, it is impossible to know it.
If you know the state and rules of the deterministic universe and use that knowledge to predict my decision, you must also in making that prediction predict your own actions, such as telling me your predictions, and my reaction to your actions. So I cannot contradict your actual prediction, unless you weren't really using the state and rules of the deterministic universe to make your prediction. (Of course I could contradict your announced prediction, if it were other than your actual prediction.)
It seems counter-intuitive that I would be unable to make a decision other than the decision you tell me I'm going to make, but if you grant that the universe is deterministic and has consistent rules, and your are able to know enough of the state and rules to make accurate local predictions, then that must be the case.
Arguing that such a universe cannot exist (or cannot be known) because it violates your intuition about free will is begging the question.
If you think the "problems" caused by the industrial revolution were prevented by WWII, you're about a century off. The industrial revolution is 1760-1820. So a hundred and twenty years before WWII.
Ps - Please don't vote, clueless voters are how we end up with Obama and Bush.
Your explanation is deeply flawed. The mistake you are making is assuming Physicalism as the zero-state, when it is clearly the more restrictive model by an extreme amount and hence would actually need extraordinary proof in its favor. What you are doing is junk-science. And hence it is a pure belief without rational basis. In fact, it also has some rather strong aspects of a mental illness as it denies individual existence.
This is hilarious. Physical reality, by scientific definition, is the "zero state" - what normal scientists call physical reality. Imagining events, though explainable through physical processes alone (see the standard model of physics), to be in some kind of magic fairyland space "outside" of physical reality is not only unscientific it also has some rather strong aspects of a mental disease as it denies reality. Once you come down off your LSD, think of a repeatable test for this other plane of existence. Show the world wrong and become more famous than Eienstien, Schrodinger, and Newton combined.
Penrose is a great physicist but just as awful about AI as hawking. Last time I checked the standard model of physics was not only deterministic it was also easily "algorithmic". So unless there is magic fairy dust from God farts making us think, it is actually possible to, in theory and eventually, simulate every atom in a human brain. It won't come to that as you don't even notice when a single neuron dies which is composed of a million billion atoms - the resolution of the simulation could be coarse and still likely work just fine. Waveform collapse happens all around us every nanosecond, there is no reason presented why meat makes it special, nor other types of noise with the same distribution aren't equally useful.
I have no idea how I fundamentally make any of my decisions. I know what argument I will come up with, but I don't know why I think of some arguments and not others. I just do. I also don't know why I will judge some argument as more "valid" than others. I just do (meaning there are "arguments" I'm not even aware I'm having). Because of that, it's obvious to me I don't have any control over my decision making process. So how can I have free will if I don't have any control over the decision process? How can I have free will if I'm not completely aware of myself?
It's completely obvious to me I don't have any kind of "free will".
Also, your hypothesis of contradicting information about your future decision is illogical. If you make a different decision, then obviously the information about your future decision was false. You do not invalidate the prediction, it's the prediction which was false because you calculated a universe where you did not give the information.
Finally, the notion of personal responsibility is only because we wrongly think we have free will and has no importance. Since I don't consider anyone has free will, I judge people on who they are (as indicated by the choices they make) instead of their "choices". In the end, the result is the same.
Why would you assume that? Taking that assertion as an axiom is in fact admitting dualism as a first principle, but there's no rational reason to accept it as a given either - I certainly don't see anything obvious about it.
Quite the contrary, it seems perfectly consistent with what we know about consciousness thanks to recent brain scanning technology.
Self-awareness is a perception that is derived from distributed brain activity, in fact it's a post-hoc rationalization of the unconscious processes that the brain performs fast and in parallel.
I don't see why that perception, like any other perceptions can't just be a result of the combined activation of our grain matter. Mind you, I don't believe that it has to come just from matter and that no other kind of "conscious substance" or "soul" can't exist; in fact I think that monism/dualism is an undecidable problem, and thus badly defined metaphysics.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
You are completely missing that any kind of perception is only something an "observer" can do. (Physical matter cannot supply an "observer", as observers can and will collapse the wave-function.) Hence you are completely missing that in your model there is nobody that has the perception of self-awareness.
Or put differently: Calling self-awareness a "perception" is correct, but it doges the question by abstracting one step. Perception requires awareness (not necessary self-awareness), and that is, again, not something physical matter is known to be able to do.
As a result, your model does only allow p-zombies (i.e. for this discussion entities that claim to be self-aware but are not), and these are not observers.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Define "non-physical". If you can't find a set of properties that can tell apart the "physical" from the "non-physical", you're just trolling (I don't know whether other slashdotters or just yourself). If your properties depend on assuming that reality is dualistic, you're doing circular reasoning.
I admit that I'm no expert with respect to dualistic theories, but I've never come across with a convincing set of such properties.
All rational require a pure unproven belief at the beginning, it's what we call an "axiom". You can't avoid those in any rational way. The best you can do is to adopt a collection of axioms that don't create too bad contradictions, and tweak your axioms when you find out that they do contradict themselves.
This is basic theory of meaning; you just can't write off other belief systems as "irrational" if they happen to be self-consistent, merely because you disagree with their adopted axioms. Doing that is not rational either.
The best thing rationality can do is assert "these are the facts we known about the world, and here we have the collection of logical models that might explain them; there's no rational way to prefer one over the others in principle when all them are consistent with the facts".
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
Sorry, sarcasm is very difficult to recognize on /.
And no, I am not. Just like many Neuro-"scientists", you mistake what an fMRI shows. There is actually no way to interpret the measurement from these and from psychological experiments reliably, as the base mechanism generating them is not understood. Think of it as making fine-grained heat measurements on a running PC. First, that gives you a massive loss of detail. You can still find the CPU (for example), but what it does is completely out of reach. And second, that PC may have a network connection that actually controls part of what it does and heat measurement cannot even detect that unless you know it is there already.
Other example: Some psychological experiments "show" that people make decisions before becoming aware of that. However there is no way to measure when people actually become aware of it, you can just ask them and assume what they say is accurate and then compare with the timing of an fMRI or the like. There is no way to correlate the time people say they became aware of the decision and the time they actually did. No self-respecting engineer or scientist would accept such data as having the accuracy needed for the claim, the measurement set-up just does not give you that.
Incidentally, I am completely fine with you thinking that you do not exist as a person. (And that is what Physicalism amounts to.) Just do not expect me to give you any respect for that messed-up fundamentalist belief.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Define "care".
If information wants to be free, why does my internet connection cost so much?
No, what physics currently think is physical reality (different from actual physical reality) is decidedly not the zero state and everything we know about it comes with proof, sometimes extraordinary proof, for example for Quantum Mechanics (because it is so very much non-intuitive). Your mistake here is that you confuse the model and the thing that is modeled. A typical novice's mistake. And of course, you cannot use the real thing as stand-in for the model, because the model explains and allows predictions, the real thing does not.
If you have a look at the history of modern physics, you will find quite a few instances where things we now accept as very likely true required decades of work to be verified well enough to be even considered possible models. We also know that we do _not_ know a lot about it. For example, Quantum Mechanics is not fully verified by a far cry. Even some relatively simple to describe problems in classical mechanics have no closed-form solutions, which means we can only approximate models for them.
Seriously, stop misusing "Science!" as religion-surrogate and learn how it works before making completely wrong statements. Actual Science makes far less grand claims than you do.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Non-physical I use here, rather obviously, for "not modeled by currently established physical theory". You confusion may stem from the difference between physical and physics. One is reality, the other is a partial model of it. The fundamental mistake physicalists make is to assume physics is the full, accurate and complete model of the physical (i.e. of reality). Physics makes no such claim at all. In fact, they are still searching for the GUT, so the physics research community is very aware that their model is limited and partial, even when you leave the questions we are currently discussing here completely out of it.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Thanks, I wanted to understand your position about "non-physical", and we have no disagreement there.
Although if you are not aiming for "non-physical" as "spiritual", I don't get why you'd describe it as non-physical. Under my model, if something belongs in the world, it is physical by definition - i.e. because both words mean the same thing - whether the phenomena can be described by the current status of the physics science or not.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
I'm not missing it (I've seen you commenting about it at another thread), it's just that I don't see how it's relevant, nor why "being an observer" is something that can't happen exclusively within the confines of the physical world as currently understood.
I find it rather funny that you think a very prominently physical phenomenon ("collapsing the wave-function") requires something from outside the realm of the modern physics paradigm.
You keep repeating that impossibility as your core argument, but I don't take it as a given ("an axiom") as you do, and it doesn't follow from the laws that explain the collapse of the wave-function. Basically you're saying "an observer cannot be physical matter because the collapse of the wave-function requires something that is not physical", but don't explain why it requires something non-physical.
It is true that current physics doesn't understand the interaction between micro and macro levels (i.e. what happens when you add what you call an "observer" to the mix), but I can't agree with your jumping to the conclusion that this interaction requires something that can't be explained with configurations of matter and energy. Under that point that you highlight as essential, an "observer" is "whatever causes a wave-function to collapse in a physical system". There is nothing requiring this "whatever" to be a form of consciousness, just that it's something external to the system studied at a quantum level.
Nothing regarding quantum physics makes me think that my consciousness requires something other than an emergent phenomenon of my physical body. If that makes me a p-zombie, I'm a p-zombie with a very rich inner life. ;-)
When I turn on the room's light, my cellphone's light sensor detects it and instructs the screen to dim the bright. You can say that my phone is aware of the environment's light.
How is that different in essence to what my brain does in my body, other than being able of much more complex processes? How do you know that what I sense as self-awareness is impossible to be the collection of those physical processes, and that I sense all of it because of - well, because I'm inside this body that is doing them?
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
Excellent. And I realize much of what I say can be mistaken as meaning "spiritual", because these people have hijacked that space. I decidedly do not mean "spiritual", that is just fairy-tale land. I still have not found a good way to make that clear from the beginning. Maybe calling it "non-spiritual" right out will work.
The reason why I say "non-physical" is because "non-physicscal" is not a word and because with your definition any non-religious (and non-spiritual) version of dualism becomes physicalism. Sure, there are valid reasons to do the definition your way, but defining "physical reality" as what physics tells us about this one universe here within the limits of what physics can observe, also has advantages. My point is just the these two things may be different and the latter may not be the full "world". As an example, while physics does have some speculative theoretical research into the possibility of multiple universes, it usually is understood as researching this one universe only, for entirely practical reasons namely being driven by observations.
Incidentally, I may be one of the very few completely non-religious and non-spiritual dualists. Bringing some kind of spiritual BS into Dualism seems just to be something most people cannot resist.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
This conversation is quite interesting :-) You're certainly a rare bird, as a non-spiritual dualist. I can empathize with that kind of dualism - being conscious certainly feels different (on a purely emotional level) to the way we see machines operate. I think I'm a physicalist mostly because of practical reasons. (And you're dead-on, my definition makes any kind of dualism a logical impossibility).
I don't rule out that there may be (i.e. exist) some phenomenon or essence that is different from the matter and energy of our current universe, which is the thing that creates conscience. But what good does it make if there's no way that we can observe it or know anything about it? That's why I said that the problem is undecidable - even if it happens to be that way, we'll never know it. Call me physical-agnostic if you will.
I try to avoid thinking too hard about such possibility because of the Occam's Razor, just as a heuristic to make reasoning about the world easier. When the only qualia ("subjective experiences") that I will ever perceive are those that I feel personally, what good does it make to think that other people also share them because of some undetectable substance? I just take it for granted that they also have them, and study the physical signs connected to those experiences - there's nothing else I can do.
Maybe you can enlighten my of why you take a different approach? I'm genuinely interested on how dualistic people reason (and feel) about conscience.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
It's not so much counter-intuitive to the notion of free will as much as it is impossible in a universe that can continue to offer any convincing illusion of choice at all.
One can easily build, for instance, an entirely deterministic system that has two possible outputs, and design it so it always outputs the opposite choice of whatever it is told to output, at which point it becomes impossible to truthfully provide as any input to that system whatever you might allegedly know the output is going be. Free will is not involved here at all. Either the universe itself is non-deterministic or else a sufficient knowledge of the state to predict the future in the first place is not knowable in the first place. If the universe is non-deterministic, then there is no reason to suggest that free will cannot exist. If the deterministic state of the universe is unknowable, then no test can be contrived which can demonstrate that we do not have free will, and suggesting that we do not have free will is meaningless, since our behavior is by all measurable standards indistinguishable from what a free-willed agent could do anyways.
File under 'M' for 'Manic ranting'
That is exactly my point.... that because we have no idea (and I suggest that we *cannot* ever know it), that your behavior is indistinguishable from that where you would have had free will, and suggesting that you don't have free will is actually a meaningless concept.
Obviously.... and again, that's my point, you *cannot* calculate the right universe in such a situation. The only conclusion is that such a situation cannot exist, but we can trivially show today that a person is capable of doing the opposite of whatever they are told to do (ask anyone who ever had a 4 year old child), so that aspect is certainly possible. The only remaining aspect of the problem is therefore whether that information about what they are going to do can be communicated to the person so it is knowable by them. If it cannot be communicated, then it cannot be known, and if it cannot be known, as far as they can ever be aware, their behavior is indistinguishable from that as if they had actually had free will anyways (and suggesting they do not have it is meaningless).
File under 'M' for 'Manic ranting'
Looks like I touched a nerve. All I said is one can take the position of Dualism to avoid your fundamental problems with Physicalism but require extraordinary evidence to actually invoke any meaningful aspects of Dualism. I guess you claim that the extraordinary evidence is already reached, but I need something more than thought experiments based on folk psychology.
So as experiments point to more and more aspects of the mind physically related to the brain you will drop more and more of your dualistic claims. Perhaps you should study more neuroscience. The split brain experiments are very interesting.
More like a child who is waiting to learn (or discover) a scientific definition of thought and intelligence before spouting off nonsense about them.
Chris Mesterharm
We very much disagree.
It would help the discussion if you'd actually give me some substance to work with instead of these grand sweeping (and generally meaningless) generalizations about AI.
AI does interesting things these days, and the applications are numerous. Collectively, AI powers civilian, military, scientific, and research in astounding ways. These applications now transverse disciplines. Some of the actions are in production today, others are in our immediate future, while admittedly, some are far off. But not that far off.
I can't even tell what you're on about. How does this relate to your point earlier about AI making decisions to benefit humankind on a high level scale? I feel like either we have different definitions of terms, or you have some wildly inaccurate misconceptions about what AI actually is.
Current (and anything in the near to medium future) AI is completely incapable of understanding those concepts. At the risk of getting into an argument of semantics, machine learning / AI isn't really capable of understanding anything. It's just running some equations on input variables that are able to be quantified and then spitting out a result. The only difference between this and a calculator is that there's a feedback loop constantly tweaking the equation.
e.g. If I could know what decisions you were making (a notion that is at least theoretically possible, if the universe were truly deterministic), I could analyze it and predict the answer you would give to a particular question, even if I told you truthfully what the answer to that question were going to be.... however, with your so-called illusion of free will, you could utilize the information that I gave you in the present about your alleged future action, and then deliberately contradict it, invalidating the prediction that I made, meaning that my knowledge about the future state was incorrect, which leads one inescapably to the conclusion that even if the universe is deterministic, it is impossible to know it.
Only conclusion from this is that you are a lousy experiment designer.
A proper experimenter wouldn't tell the person about the prediction about his decision, he would escrow the prediction. A crude example is write it in a paper while not showing to the experiment subject, handing it to him in envelope while he makes the decision, and THEN opens the envelope.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
If the future is predetermined, why should whether or not the person knows about the prediction beforehand affect the outcome? Indeed, if the future is predetermined, it cannot. If the information cannot be communicated to that person, it is unknowable by that person at the time that they appear to make a decision, and therefore indistinguishable to that person from free will.
File under 'M' for 'Manic ranting'
While there is some disagreement on the meaning of Physicalism it certainly doesn't include thinking the current laws of physics are complete. Who would make such a ludacrious claim.
Chris Mesterharm
Both your points are illogical.
Your first point is that because we are not conscious of ourselves, it follows that we have free will. That's not the definition of free will. After all, if we follow your logic, it would mean that a computer program has free will because it's not conscious of itself.
It's not suggesting that I don't have free will which is a meaningless concept, it's the concept of free will itself which is meaningless as it implies non-causality.
Your second point is that we cannot calculate a possible universe because we could modify it after the calculation. Again, this is like saying we cannot calculate x+3 simply because we could change the value of "x" after finding the answer.
If the " future " is predetermined, there is no choice "whether" to tell him about the prediction. But if only the subject's decision part of future is predetermined , it is a function of other events leading up to the decision. I.e. what is predetermined is a function taking precursors as input and giving the decision as output.
So telling him about the prediction assuming he was not told about the prediction can very well change the future decision and may not match the prediction.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
Maybe you can enlighten my of why you take a different approach? I'm genuinely interested on how dualistic people reason (and feel) about conscience.
Well, originally it may have been simple terror-management in an established way, namely I believe that some part of me is not tied to this mortal shell. As religion never did anything for me at all and I saw right through it even as a teenager, I eventually happened on the idea of dualism and it made a lot of sense to me.
At that time, I would probably have stopped caring, where it not for this AI panic we have had going on for a few years. It got me to think on the nature of intelligence. From my education and work, it is pretty clear that there will not be anything resembling humans with respect to capability to understand and it is completely clear there will never be any Turing-machine type computer that has self-awareness. (Let's not discuss that now, let's just say that for my own stance on the issue I have reason to be very sure about this, with at best a tiny chance on me being wrong.) That got me to think on dualism again, because how do humans do insight and self-awareness? A quantum computer is a Turing-machine as well, just one with unusual performance characteristics, so that is not the answer even if Quantum Computers should eventually work (a big if).
So what then? The hypothesis I finally formed is based on this: There are a _lot_ of quantum effects with random behavior in the synapses of the brain. If some external entity could influence those probabilities even by a tiny faction, that would be a pretty powerful communication interface right there and a way to attach an extra-physical part of a person. Of course, we have not measured such tiny deviations in probabilities. But at the same time, this may require a precision that practical physics in this space has not yet achieved and may be hard-pressed to ever do so. Keep in mind that dynamic analog measurements become very dicey somewhere around 14-20 bits of resolution, and that is not so much, hence respective measurements may be infeasible in practice if the effect is small enough. The other thing may be that this effect only manifests itself if a quantum-randomness using device gets to a complexity level where the "interface" becomes good enough for a view into this physical world. That may be a decision or choice on the extra-physical part of a person, i.e. to not attach unless it works well enough. There may also be some rather strong distance limitations to this effect (we have no telepathy, after all) and hence all these random quantum effects may have to be packed into a small volume for this to work.
Now, that model is not in opposition to what Science currently says about quantum physics if the effect on individual random quantum effects is small enough. Physical models have proven to be not quite exact time and again. It does solve the question where intelligence, understanding and self-awareness can come from in a physical world, namely from outside. This removes the need for any "magic" "emergent properties" that matter would have to have in order to be able to create them. It also solves the problem how CS does not even seem to have a theory how intelligence could be created artificially, despite about 40-60 years if intense research. No other practical CS problem has proven that hard. Also, CS has absolutely nothing on self-awareness.
The beauty of this is that it works completely without unfounded (and often created to manipulate) religious or spiritual ideas. This model merely says that there is obviously some rather prominent possible interface in a human brain and something is attaching itself to that interface. It is also compatible with some other models of reality, for example the idea that this universe is a simulation, which has some things going for it. It would possibly allow AI (if the quantum-interface can be packed tightly enough), but that would come with intelligence, consciousness and free will as well. (
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Well, the thing is that current Quantum Theory is silent on the matter. It basically says that in some step from microscopic to macroscopic something magic happens and the macroscopic object then is an observer. That is however not consistent with what we have in theory as this would say Quantum Theory does not scale. The theory itself does not have that limit, but the reality it tries to model apparently has it. So either we get rid of the idea of an observer completely, or we accept that quantum theory is a rather incomplete description of how things are. Most physicists tend to do the second, and I agree.
As to "awareness", your light sensor has no awareness of there being light or not, it does not understand what light is. In a sense, it has a "reflex", but that is it. The same is true for the rest of that simple regulation circuit. Actual awareness can produce insight. For example, you cellphone only knows to dim the lights, it does not know why or that there apparently is a light-source. It merely executes an order literally it has been given before by its programming. It has no way of leaving the box it is in.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
The "Halting Problem" is mathemetician's finger twiddling, and has nothing to do with real Engineering...
I may be one of the very few completely non-religious and non-spiritual dualists
If your beliefs regarding dualism can be expressed as "maybe there is more to consciousness than can be explained through current physics"
then your ideas can reasonably be described as "non-spiritual". But that's clearly not the case.
The beliefs that you have expressed in this post (and numerous others in the past) can be better described as
"there just has to be more to consciousness than physics. Why? Because you can't prove it isn't so!".
Identifying yourself with any kind of "ism" is foolish in the absence of supporting evidence, and you have none.
He based his argument on a novel but quite defensible interpretation of Godel's Incompleteness Theorem
If you actually read this "interpretation" it says in effect "if we assume that people can deal with incompleteness and computers can't,
then we can conclude that computers will never be as intelligent as humans".
Sure, but since he never gives the slightest reason to accept that premise, the whole concept falls apart.
Well, you seem to feel threatened by my ideas, which you sum up incorrectly. Apparently you have not understood a single word I said. What you do is not criticism, it is propaganda. Pathetic. This reaction is however entirely typical for somebody that sees his beliefs challenged and knows he has no good answers.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
It seems that our only core disagreement is regarding whether awareness can emerge from a combination of the physical that we already know about, or whether it requires a yet undiscovered phenomenon (what you call "extra-physical") that would influence them.
There's nothing "magical" in emergent properties of complex systems being able to adapt to their environment in intelligent ways, or producing elaborate results; the only thing that seems magical is how such complex behaviors can appear out of very simple, unrelated rules - but it can be seen happening everywhere, if you know where to look. See for example how Conway's Game of Life produces persistent, moving structures out of single-cell activation rules, or how the parable of the polygons gives raise to segregation from individual decisions that are not particularly racist.
Such emergent behaviors may indeed be influenced an react to an external environment, in special when the survival of the whole system depends on it (have you seen ant colonies fighting an aggressor?) - survival of the fittest means that only systems able to adapt will be around for long; so the well-tuned systems are self-selected, and those without the "magic" property will disappear. So this model too works without unfounded ideas, just like yours; and it doesn't require any "undiscovered particle", while your model requires postulating a sort of "Higgs boson of awareness" that we don't nothing about.
As for CS not having a theory for how intelligence appears, that is be true, but neuropsychology does have a good basis for one; as I mentioned, our conscious process -when scanned through magnetic resonance- looks a lot like creating post-hoc rationalizations of the reflex analysis performed by the lower brain functions. That model could be reasonably expected to provide a basis for true artificial intelligence in the future.
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
Oh, and I forgot - emergent behaviors of complex systems are not subject to the limitations of Turing-machines completeness; such behaviors are not described as symbols within the system, so they cannot be "diagonalized" (which is the basis for the limitations of formal systems such as Turing machines halting problem and the GÃdel's theorem in mathematics).
We don't know whether the "emergent computation" of distributed automata systems is more powerful* than Turing's computable functions, but we have no reason to believe that they are equivalent. A systems that performs computation through emergent adaptations rather than manipulation of symbols might be strictly more powerful.
*(We know it's not less powerful thank's to Rule 110 being Turing-complete; anything a symbolic computer can do, a cellular automaton can do as well).
Singularity: a belief in the "God" idea with the "demiurge" relation inverted.