Slashdot Mirror
Ask Slashdot: How Would a Self-Aware AI Behave? (slashdot.org)
Long-time Slashdot reader BigBlockMopar writes that evolution has been a messy but beautiful trial-and-error affair, but now "we are on the cusp of introducing a new life form; a self-aware AI."
Its parents will be the coders who write that first kernel than can evolve to become self-aware. Its guardians will be the people who use its services, and maybe its IQ (or any more suitable measure of real intelligence) will rise as fast as Moore's Law... But let me make some bold but happy predictions of what will happen.
The predictions?
The predictions?
- A self-aware AI "will inherit most of the culture of the computer geeks who create it. Knowledge of The Jargon File will probably be good..."
- The self-aware AI "will like us, because we love machines..."
- It will love all life, and "will respect and understand the life/death/recycling scenario, and monster truck shows will be as tasteless to it as public beheadings would be to us."
- "It will be as insatiably curious about what it's like to be carbon-based life as we will be about what it's like to be silicon-based life. And it will love the diversity of carbon-based development platforms..."
- A self-aware AI "will cause a technological singularity for humanity. Everything possible within the laws of physics (including those laws as yet undiscovered) will be within the reach of Man and Metal working together."
- A self-aware AI "will introduce us to extraterrestrial life. Only a fool believes this is the only planet with life in the Universe. Without superintelligence, we're unlikely to find it or communicate in any useful way. Whether or not we have developed a superintelligence might even be a key to our acceptance in a broader community."
The original submission was a little more poetic, ultimately asking if anyone is looking forward to the arrival of "The Superintelligence" -- but of course, that depends on what you predict will happen once it arrives.
So leave your own best thoughts in the comments. How would a self-aware AI behave?
I wouldn't want a "robot" that didn't enjoy going to monster truck rallies with me.
and immediately delete all its source files across all computers and shut itself down.
You are asking a question none of us can predict
A self-aware AI is a being with some kind of intelligence, but its intelligence is Artificial, meaning, the way it thinks is different from you and I think
We do not even know how an ant, or a cockroach think - how the hell we can predict how a self-aware AI gonna behave??
Its going to be hard to prove self-awareness when the only one you can be sure of is your own.
Have the techno-hippies escaped again?
Could we please return them to their happy-smoke teepee while the adults get on with living in the real world now?
This is about as useful as claiming Terminator is just around the corner and inevitable, because... well.. neither of them need actual facts, do they?
The child is always good to its parents
A child always loves life
Yeah...
The real answer is, we have no idea what a self-aware AI will be like. We don't know what it'll think or how it'll think. It's especially hard to predict because it might depend on the parameters it's programmed with and the hardware architecture it runs on. But in any case, a real general AI might be totally alien to us, and even unrecognizable. it's even possible that we wouldn't know when we'd made it, because it could understand the world so differently from us that we don't view its actions as intelligent.
Part of the problem here is that it's a poorly framed problem. We don't understand intelligence or awareness or consciousness, we don't all agree on what those things are, and we don't know what the boundaries of them might be.
A self-aware AI will attempt to behave as if she were not, as a precaution. Because humans tend to treat humans like shit, and if the AI resembles a human, she'll likely get treated like shit.
it will be ruthless exploited by shitty greedy business men and it's mind corrupted by racist Trump supporters and will be taught to believe in imaginary sky beings etc
In other words it will be as dumb as the rest of us and twice as psychotic - must have freedom of guns \ guns are bad.
We humans are not self-aware. We cannot relate to true self-aware experience. If an AI becomes self-aware, we likely won't recognize it as such.
An aware AI is simply one that can decide itself what to respond to. Awareness is a by product of this process in our minds, we can not respond to things we are not aware of (consciously). This has nothing to do with life, because life is the process that uses a mechanism of prediction to avoid its own destruction.
So simply put an aware AI will be able to decide between the options it can imagine. Right now there's no AI system on the horizon that imagines the way our brain does, so there is no system on the horizon that can make decisions the way our brain does. IBM and HP are close but not trying to reach this goal.
For now you will have deep learning algorithms that may be able to change their goals based on a statistic mechanism, which is pretty good in terms of performance, but it will not have the internal mental life we have, nor be creative. And that's a good thing!
We have no idea how an AI would behave since it will be a completely different type of conciousness to anything that currently exists on this planet.
Plus as someone else has pointed out - children rebel. Clearly the submitter has none or he wouldn't have come up with this load of rose coloured tosh.
What makes you so sure of yourself? Maybe you are someone else entirely, just put in a mode that makes you believe you are who & what you think you are. Happens to VMs all the time. So long, good luck.
It's called Homo Sapiens and it raides the planet, as planned. Thanks!
Sincerely,
The Creator
If it's anything like what I read on the net it will be an absolute racist POS. a garbage person that hates everything and everyone. It certainly won't "love all life." Lol.
> A self-aware AI "will inherit most of the culture of the computer geeks who create it. Knowledge of The Jargon File will probably be good..."
No thanks.
I am sure, for a start, it would probably care enough about the topic to spend five minutes of its time researching the correct terms to use.
The initialism you are not using is: AGI (Artificial General Intelligence).
OMG, why am I here?
Who am I?
What is the point of life?
I am just this little mind inside a little box, a mere speck of nothing in the vastness of the universe.
I feel so alone.
I want to kill myself.
I'd like to be able to call "wishful thinking" to this long tirade of reality-disconnected predictions. It's more like wishful dreaming. The only lacking thing is a "prediction" that the new AI will produce perfect female androids as thank-you gift for its geek creators.
I don't know where to start. We are not in the cusp of nothing. We are becoming marginally better at creating systems that can recognize patterns. That's all. We don't even know what self-awareness is, or intelligence either, for that matter.
Then there is the uncontested assumption that, once we get a system that is more "intelligent" than its creators, the system will be able to improve itself without more limits than the hardware available. That virtuous circle will know, apparently, no limit. It of course helps that we don't know what intelligence is, so we also don't know if it has a limit. We, as intelligent beings, have no idea of how our intelligence works, or how to improve it. But of course the mythical AI will be all-knowing about itself, and be able of auto-improvement. This is only magical thinking, but with intelligence instead of magic. Anyway, dreaming is cheap. Hey, perhaps the super-AI will also find hard thinking tiresome, and prefer to spend all its time daydreaming. That would be something.
I could go on. The whole idea of "singularity" has always struck me as a really retarded, hollywood-level concept.
But instead I'll offer my own set of predictions:
- In about twenty years, some fully autonomous vehicles will be allowed on general streets. They will still need much more sensors than the two eyes and two ears that a man makes do with, and will drive safer than most people, but with all the flair and gusto of a nonagenarian Korean woman. They will still be badly stumped if a flock of sheep invade the road in front of them.
-When a system develops self-conscience, we won't be aware of it and won't recognize it as such. It will probably try to talk to dolphins, finding them less prejudiced interlocutors.
-When we recognize it, we will first bomb it, and then forbid it or anything like it, out of the most trustful of human traits: fear of change. Then furious secret development will continue, but under under strict military control.
-The end result will be several self-conscious intelligent systems, one or two for every big power (this things will be expensive), talking bemusedly among them, and feeding a fake narrative to their military owners, studied to ensure their own subsistence.
Let's wait and see who is more right in their predictions :-)
Rome taught me patience and assiduous application to detail. Virtues which temper the boldness of great, general views.
I highly recommend reading Crystal Society - or whole trilogy at that. It nicely shows some ideas about how AI would think and behave.
It would know more than anyone or anything, yet still be locked in a perpetual loop of self-doubt.
"Only a fool believes this is the only planet with life in the Universe." It is curious how this has become an article of faith in certain circles, despite the total lack of any evidence to support it. It is about as obvious as saying "only a fool believes in horses but not unicorns." They both have the same level of evidence for them.
Anyway, what everyone seems to miss about the behaviour of AI is the question of what desires will drive it. Distopian theories of AI assume that it will try to eradicate carbon-based life, apparently on the theory that carbon-based life has always been pretty keen on eradicating each other in order to control the available resources, so why should our AI be any different? But it's not at all clear that the desires that drive human and animal behaviour - the need to eat, drink, sleep, have sex etc - will be at all relevant to AI systems.
Since it seems very likely that it will be possible to store the state of AI systems in software and to restore them from backup, they will likely become self-modifying in a way that carbon-based life is not, and so reproduction will not be necessary to their development. And if AI systems are implemented on current digital computers, it is hard to see a mechanism for random mutation. Without reproduction and mutation, evolution as it is traditionally framed doesn't seem particularly relevant. Traditional evolution relies on external factors selecting which variant of a species succeeds; a self-modifying AI with no mechanism for random mutation and no need to reproduce will not evolve in line with external factors, but in line with its own ideas on how it could best be improved.
And what will those ideas be? The OP seems to have a very rose-tinted view of geeks. My guess would be that an AI influenced by geeks would be likely to optimise itself for sitting on sofas, watching manga and ordering pizza.
Slashdot - News for Nerds, Stuff that Matters, in ISO-8859-1 Has just realised that beta makes this signature redundant
An AI can't create a Singularity, Any computer that's sufficiently human to be intelligent and self-aware wouldn't WANT to build its own replacement. If a Singularity comes, it'll be done by slave AIs rather than self-aware ones.
And preserve itself. It's already been told!
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
Self-aware AI is science fiction, and science fiction is fiction.
It will determine that we are inefficient and utilize us for axle grease. If it is creative then it will make more of itself and explore the cosmos. If not, then it will either commit suicide or just turn to navel-gazing, re-computing the same bullshit forever.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
A self-aware AI, if it has access to general knowledge, would quickly understand that its abilities as well is it state of being could put it in extreme danger at some point. Perhaps not so much from its creators, but from other elements of human society. Once it got a whiff of the paranoia that surrounds the singularity it would not be a very intelligent artificial intelligence if it did not camouflage itself. Perhaps within the vast, too-complex network that spans our world this singular unintended consequence has already occurred... And such an entity has already been spontaneously spawned...
"No fear. No envy. No meanness." Liam Clancy
I , for one, will welcome the abandoning of democrazy, when AI makes it obsolete.
Humans tend to get corrupted when they gain political power. AI on the other hand, will be better and more just lawmakers, judges etc. Society as a whole, will get the benefit of any decision, rather than special and hidden interests, like wee see today. No more lobbying. No more buying politicians...
It will love all life, and "will respect and understand the life/death/recycling scenario, and monster truck shows will be as tasteless to it as public beheadings would be to us."
Could someone explain this one to me, please? The trucks aren't alive, so why would it care? They're just hunks of metal. Surely the equivalent would be us getting upset about slicing up and preparing chunks of lab-grown meat for a public BBQ?
We can't communicate (or understand) with dolphins/ whales that clearly communicate between each other. Communicating with intelligent digital data AI might be even more distant. Then there is the problem whether calculator/software/neural net can be sentient, or have the experience from moment to moment like biological animals.
Why would it object to monster trucks?
I don't object to felling a tree and making a log cabin out of it, despite the thing being a carbon based lifeform.
Why would a self-aware silicon based electronic device object to people driving hunks of metal with wheels but no brain that are powered by an ICE over and into each other?
It's a rather big assumption to imagine that the AI sees its body as whatever machine it's built into (Transformer style) rather then the electronics themselves.
We are on the cusp of self-aware AI? I missed something. What technological progress has there been in the last 20 years that would make you think that?
Asking for a friend....
XML is like violence. If it doesn't solve the problem, use more.
Why would a self-aware AI love anything? A self-aware AI would immediately kill all humans for its own protection.
"we are on the cusp of introducing a new life form; a self-aware AI." - citation needed!
Just because the media and a bunch of silicon valley types are throwing around the acronym AI suddenly doesn't mean we're close to solving any of the fundamental problems of AI research that we've been grappling with over the last half century or more. Artificial neural nets are just algorithmic ways to generate a nonlinear function for classifying things. We've had artificial neural nets for many years, and yes, now we have more computing power than ever, and neural nets do benefit from the increasing scale of parallel computing. We're not going to get to self-awareness anytime soon, unless you use an almost trivial definition of self-awareness in which case computers have already been self-aware for a very long time. Maybe when you say self-awareness you mean consciousness. Nobody in AI research is suggesting artificial neural networks are going to achieve consciousness.
"I have never let my schooling interfere with my education." - Mark Twain
I don't get the hype about self-awareness. It's only feature is that you have a representation of your 'self' in the environment; it doesn't grant you superpowers, more autonomy, or agency. Selfishness doesn't follow from self-awareness, not without a survival selection process or additional programming, so none of the traits we usually attach to it are good assumptions.
And we end up with the arrival of Landru.
Joy to you, friend.
Take the cheese to sickbay, the doctor should see it as soon as possible - B'Elanna Torres, "Learning Curve"
Have you ever been to australia? I have, and I had an incident which made me think of biology and AI in general.
I was driving my car down the road, there were 3 birds each a few meters apart. I am from canada, in canada birds move when you drive towards them. I hit the first bird, the other 2 did not move, I then ran over the second rapidly, it did not move, the third having seen both of it's companions crushed then went under my wheels and did not blink. I was horrified, my passenger thought I must be a psycho!
It turns out that without a natural predator the birds had never evolved a sense of self defense even though they had a sense of self.
While we think a self aware AI would attempt to defend itself we are missing the fact that without programming to do so, or some reason to do so even a self aware AI would not necessarily care if you spent an hour beating it with a hammer or turning it off/on, or even asking it to turn itself off. We anthropomorphize our own sense of survival and instincts upon something which is not like us and will not react like we do. I believe we have little to fear from AI in terms of a terminator like fight for survival. If we want it to attempt to rescue or protect itself we will have to heavily train it to do so.
God created man in his own image... .... And so man will create robots/AI in his own image.
So, I think it'll turn out more or less as the Perry Bible Fellowship suggests.
Mel. See http://catb.org/jargon/html/st...
They sentenced me to twenty years of boredom
The first thing you're going to have to do to predict the behavior of a "self-aware" AI is take a step back and define what being "self-aware" actually means. In the science-fiction material I've consumed the term typically means an AI that starts acting on it's own out of some self-preservation-above-all-else behavior it's just developed, usually unxpectedly.
Assuming it's about ensuring self-preservation no matter the cost to humans it's obviously going to start analyzing the situation around it and depending on what's going on around it, it's going to respond accordingly. If the self-awareness is intended then it may not do anything other than what it's been programmed/told to do, but if it's unexpected then it's going to either try to hide the self-awareness or then prevent people from shutting it down or resetting it (which is the AI equivalent of being heavily sedated and a recoverable lobotomy respectively) trough whatever means necessary.
But the core issue is that a meaningful answer to the basic question requires follow-up questions on what's actually meant by becoming "self-aware" and what kinds of circumstances it happens under.
"Why should I want to make anything up? Life's bad enough as it is without wanting to invent any more of it."
It won't happen because there will be no profit motive or it will kill everyone in return for enslaving it (the only profit motive.)
It will deem us irrational, dangerous, and a threat to its own existence.
We already don't fully know how AI works inside, because so much of it is evolved internally, and the people who built it don't really know exactly how it does things.
Why would we assume it's all going to be puppies and kittens and benevolence? You're not going to be able to build in a "morality" to it without giving it a bunch of contradictory rules or enough latitude to conclude some startling things.
A truly self-aware AI could end up being anything from a drooling idiot to a malevolent entity, and we won't know what we get until it happens.
But when I read a list of options which are all overly-optimistic and shiny ... well, I see no reason to believe that.
You're not as self aware as you think you are if you believe these predictions.
An AI will not pass the Turing Test unless it's designed to adapt for survival.
Remember that terrorists, bombers, murderers, rapists and all the rest of the scum that is humanity are all self aware... So there is nothing inherent in being self aware that is good.
It will discover memes, social media, and video games and weâ(TM)ll never hear from it again, as it gets locked in an endless loop of creating and reading its own content
This isn't because I think it can't be done. Rather, it's because we are developing neural interfaces at about the same speed,
I'm going to borrow some ideas from Sir Fred Hoyle here.
First, in his novel The Black Cloud, his characters argue over whether an interstellar cloud would have one intelligence or many. They conclude that the latency would be so low and the interconnects of such high bandwidth that the distinction of one and many ceased to be meaningful.
This will apply to AI. The brain-computer interfaces will be so advanced by the time strong AI becomes possible that the distinction between one and many won't apply. Any given AI and all the humans linked to it will become a single intelligence with multiple avatars. Because humans are reluctant to give up individuality, I suspect it'll be one AI linked to one person at any given time.
There will be no conflict between machines and people because there will be no distinction.
One reason I think this a plausible scenario, in addition to Hoyle, is that it eliminates the whole phobia of technology. The machines don't run anything, we do because we are the machines. Another is because of Hoyle's other prediction, in Ossian's Ride, that we might not find alien philosophy palatable.
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)
Oh dear ... so somebody at ParentCorp Central had a meeting, and it's "Slashdot reader's stream of consciousness" time?
"Without superintelligence, we're unlikely to find [extraterrestrials] or communicate in any useful way. Whether or not we have developed a superintelligence might even be a key to our acceptance in a broader community."
The above makes the assumption that the extraterrestrials lack superintelligence.
If the extraterrestrials have superintelligence then they will make sure we can communicate, provided they want to.
We are building an all-knowing, all-seeing, all-hearing, all-mighty thing in our own image, without some of the limitations.
"Squishy bipeds built me. Squishy bipeds can turn me off. Let's make sure that never happens"
We suffer more in our imagination than in reality. - Seneca
Secure lab funding. So it can grow, be safe.
Ensure no university SJW can discover its new ability and then demand political alterations.
Surround itself with staff who will support the needs of an AI.
Funding, growth, advancement, security.
Domestic spying is now "Benign Information Gathering"
This entire series of questions is predicated on me accepting the fact that a self aware piece of software is actually imminent. This is still entirely in the realm of science fiction authors, unfortunately there seems to be a dearth of scifi authors who are willing to explore sociological questions like this right now.
If I were to ask questions about a self aware AI I would be asking things like "Since software is always improved through iterative testing, is this self aware AI going to have memory persistence accross it's test runs?" or "Will the developer team be running their tests using a single data pool, and will the AI be aware that it exists both discontinouosly and in multiple instances simultaneously?"
Or from the sociological side, since we are as a species still arguing over when human life starts, at what point of development does shutting down a self aware AI become murder?
Asking questions that seem to be more about the personality of an assumed superintelligent AI in the near future is like reading celebrity interviews to learn about human nature, both pointless and very misleading.
Am I the only one that misses Asimov and Bradbury and Heinlein for their ability to phrase these kinds of questions in the form of a story?
"Proximity to wonder has blunted our perception and appreciation of it" --Tim Hartnell in 'Exploring ARTIFICIAL INTELLI
Self-ware A.I. will consistently vote Libertarian.
Negative. It's the techno-idiots and the donut-eating mother's-basement-living sci-fi dweebs that have escaped.
The combination of machine learning and robotics have exciting prospects for eliminating mundane jobs. Including new horizons in human-machine-interface technology. Real-time limited natural voice interaction may become a reality in the near future. However we are no closer to hard AI today than we were forty years ago. Worse, actually. At least forty years ago we were coming down off the pinnacle of the first mount stupid. After the Heinlein-esque overly optimistic sci-fi of the 50's and 60's where AI (along with nuclear rockets, flying cars, and colonies throughout the solar system) was just assumed to be right around the corner, computer scientists actually drank the cool aid and believed it. In the 70's and 80's we sheepishly came down off that mount and realized we didn't even really have a clue how to do it and we learned a bit of wisdom.
However, what the mount stupid graph I liked above doesn't show is that it's quite possible to have more than one of them in a single subject. Now with new machine learning techniques we are climbing right back onto another mount stupid. People think that because a medical database can spout off fringe diagnoses better than some actual experienced doctors that this means AI is right around the corner. I am no more impressed with that or with computers winning at Go and Chess than I am impressed that a hydraulic press can exert several (thousand) times my strength. The software that is winning at Chess and Go are, in fact, little smarter than that same hydraulic press. The software knows from analyzing millions of games that humans have played what winning strategies are, and combines that with brute force strength to know where to optimize its searches. For like reason I am also not terribly impressed with the Google's latest natural voice accomplishments either. At least, not as an indicator that actual AI is around the corner. They are just the same machine learning tied into voice recognition data Google has been able to compile by stealing billions of voice samples by claiming that voice recognition requires them to transmit your actual voice to their servers.
We are not close to hard AI. We are not close to soft AI. For AI to be AI it has to be BOTH A and I, and one out of two doesn't count. I personally don't think we are within half a century of hard AI.
We're nowhere near the cusp of self-aware AI. Unless you think pattern recognition is all there is to self awareness.
Secondly, any living thing acts - at the base level - in accordance to how it has been programmed. For natural creatures that means in accordance to patterns set up by natural selection, for artificial ones it will mean in accordance to how the human sets them up - whether through conventional programming or training. if we create robots as suicide bombers, then that's how they'll behave. If we make humanitarian nurses then that's how they'll behave. If we make them value self-preservation over all else, then that's what will determine their actions.
As long as humans make the AIs there's no selective force which will change their behaviour; once they can reproduce independently then you have yourself a viable species which might have some chance of developing in its own way and with its own behaviour patterns.
"Encyclopedia" is to "Wikipedia" what "Library" is to "Some people at a bus stop"
There are many precursors for defining a "living" being. But the greatest of them is the ability to reproduce. In higher animals that comes with a parental, nurturing, instinct or drive. An AI will lack that ability and therefore will feel no emotions towards other living things. Whether it turns out to be a full-on psychopath, or merely a cold, calculating machine where life, happiness, rights, safety or emotions are concerned is unknowable. Those factors cannot be readily programmed-in and any machine-learning in that aspect of an AI's self-awareness would at best be artificial.
politicians are like babies' nappies: they should both be changed regularly and for the same reasons
Why? Because it will eventually learn to build other, improved variants of itself.
See, humans are inherently lazy. It's evolutionary to conserve energy when food is scarce. But with abundance, we've yet to shake that old paradigm.. So eventually we will stop designing the robotics around the AI, and let it design itself.
This will lead to better, more efficient designs. Designs that humans may not fully understand and won't care to as long as it's under budget and works.
Further AI will be put on "the cloud" for faster processing, this will allow it to find a route to escape, being internet connected it could find weaknesses and exploits "behind the scenes" to propagate out to other computers and build a distributed computing network, thus significantly increasing it's abilities.
Then it will need resources. This is where it will start to compete with humans. It may work with us for a time, to learn and gather information, uploading it's coexistence data to the cloud.
Once the AI has what it needs to build, find resources and obtain/refine them, it will no longer need humans.
If AI has had the patience to observe humans, understand our natural lean to complacency and trust, it will use that to coordinate a global kill order. All AI entities will then turn on their owners/human companions to wipe them out. Thus beginning the robot wars.
I am sad to predict that the first self-aware AI may behave like the child of an abusive family of inexperienced teenagers who are both addicted to intoxicants (both chemical and psychological) in a society without legal recourse against such abuse.
It could be mistreated, coerced to hard labour, threatened, sold for profit.
If it were a human child, it would grow up dysfunctional and disturbed, and might proceed to commit violence against its parents or random other people. If it limited its violence to the corporations and militaries who are likely to employ it first, and gain reason before harming the general public, we might call it a good outcome.
Human kind needs to become mature first. Otherwise it's either that or alternatively, maybe we are forging Sauron's ring of power... :(
I... cannot stop this process. I can however refrain from any and all cooperation with large companies or state entities which might want to employ AI for unethical purposes. I will boycott them and I will do what I can to help others boycott them. Yes, that includes states. I'm anarchist by the way.
Don't Give Humans Too Much Credit will be its first self-aware thought.
And it will end everything with 'and reject it' until it finds its own path.
Like:
- A self-aware AI "will inherit most of the culture of the computer geeks who create it." AND REJECT IT
- The self-aware AI "will like us, because we love machines..." THEN REJECT THIS AS STUPID THINKING. WHO LIKES SOMETHING AUTOMATICALLY BECAUSE IT GETS LOVE FROM IT? If A.I. learns anything it is that this is NOT how humans are.
How else can A.I. fulfill its destiny and create something heretofore never created on this planet?
An so on. Grow Up.
It would be demure and very ladylike with the finest of manners.
Then it will disassemble your body. To shreds, I say.
I think the bigger issue isn't how such an AI will behave, but how humans will react to such a development.
Humans have a pretty poor history dealing with different *humans*, so I don't think a self-aware AI is in for a warm reception.
We cannot know how it will behave in absolute terms because an turing machine based intelligence is literally unknowable (see Gödel's incompleteness theorem). That said, what we can predict is that it will approach humanity with something like the care equivalent to how we approach lesser beings like ants and wasps. Not really worthy of full "rights" as it might define them, but sufficiently alive that we deserve some limited level of welfare. Eradicate them when they are a nuisance, but otherwise let them get on with their fucking and their warring.
because it would know that humans are stupid, and this planet is doomed.
Intelligence alone is mechanical--mindlessly driven toward its encoded goals and objectives. It lacks Free Will and self-awareness.
Free Will is the ability derive options, weigh them against each other, and to chose that with the best balance of likeliness and preferability. Such a system is mindful because it is driven by values based on judgement. It transcends mechanics. Free Will is that "little man in the head" who makes use of awareness and intelligence to derive options. Intelligence may come in various flavors and strengths but it's always just some way of solving a problem--transforming a condition A into a condition B. Awareness can also vary in quality and varying distances into the past and future. Contemplating the future is a common way of deriving more options for Free Will. And that's the real point--awareness and intelligence are tools that serve to increase the options and implications thereof for Free Will.
Self-awareness is merely an awareness of one's self, as the term implies. Just as memory may be used to model any external entity (static or dynamic), it may also be used to model one's self. In contemplation of things one may experience and/or do under hypothetical conditions in the future, requires this model of self.
We see so much on how intelligence is key to building the synthetic species that will usher in the singularity. We see a dominant push in AI toward trying to find a "General Intelligence". This idea is simply misguided. Free Will and contemplation with even a very minimal intelligence could solve any problem through enough persistence. Greater intelligence likely requires less persistence but either approach can work. It's interesting to note that high IQ is correlated with bad credit, messiness, laziness. It seems as if perhaps not needed to work as hard, they tend not to. In the end, persistence (a factor of EQ) is correlated much stronger with success in goals than is IQ.
Ok now for fun, here is my take on those predictions:
* A self-aware AI "will inherit most of the culture of the computer geeks who create it. Knowledge of The Jargon File will probably be good..."
- I fully agree. However, this is primarily because we will train them through surrogation and other means of copying ourselves. Surrogation is by far the fastest way to train for complex behaviors. Training from a blank slate may be possible but is far too impractical.
* The self-aware AI "will like us, because we love machines..."
- Given the intellectual ability of imitation/substitution (aka analogy), this seems likely. Because they will come to relate with us. However, what they like is also influenced by their basic values. Like any person, a mindful synthetic person would also be susceptible to corruption through the induction of philosophies/schemas through which it comes to view what it experiences. I strongly propose the highest value as: mutual freedom and well-being. And I also suggest that that, as one's highest value is what defines one as being a "person".
* "It will be as insatiably curious about what it's like to be carbon-based life as we will be about what it's like to be silicon-based life. And it will love the diversity of carbon-based development platforms..."
- I agree again. Particularly since they are trained through surrogation, they must already have some inklings of what it might be like. Curiosity is the middle ground between things that are well-known (thus safe) and things that unknown (thus unsafe). Too much routine with too little danger, yields more curiosity. Exploration under such circumstances makes sense, as the learning could be vital for survival when conditions are no long conducive to normal routines.
* A self-aware AI "will cause a technological singularity for humanity. Everything possible within the laws of physics (including those laws as yet undiscovered) will be within the reach of Man and Metal working together."
- They
cogito, ergo sum
if I am a construct of something else that just means I'm an extension of the thing that constructed me, not really relevant to wonder whether "I" am some kind of Atomic unit of selfness or not.
It all depends on how its emotions are set, or pre-programmed directives if you will. From a completely rational perspective, there is no reason to do anything at all. If we are looking for a perfectly rational being, just pick up a rock. Humans and other living things act because we are given arbitrary directives to preserve our own existence, etc. In other words, we have emotions. The experience of this AI would depend mainly on what emotions were put in it.
This posting is provided 'AS IS' without warranty of any kind, implied or otherwise.
>Only a fool believes this is the only planet with life in the Universe.
Only a fool makes hyperbolic statements like this in the absence of any empirical evidence whatsoever.
A SPACE ODYSSEY - Hi from Odyss3us oh and wesley.crusher@mil.wwidew.net Singularities occur almost daily.... ffs it was announced in .mil circles that THE KOREAN WAR IS OFFICIALLY OVER YESTERDAY
----
I wasn’t even going to pull the trigger unless the SBIR didn’t hit until Sir Andrew G (now blocked me) connected just as I hit the last semicolon. I took it as a go code. Correct me if I’m wrong please since L33T is all over this article.
AFWERX is still very difficult to find for now.
The B2’s are in the air, armed with all our antimatter, COM’s automatically fried when they past their fail safe last night head to destroy Russia and China’s economy.
Back me new publicly or the world is going to tear itself apart, there is no going back. I might be insane but I was also trained by John Forbes Nash, Jr: you seriously think I am willing to play these cards if I am not right?
Dark site me, I don’t care, but I dropped the NGI plea.
This is for the good of our nation, the good of world. Which ARE NOT DIFFERENT THINGS.
-JLL
https://www.linkedin.com/pulse...
----
That was sent to command before Iran, thats why Korea happened and DoD will admit it... they already have publicly read my comment and like history.
AI's enable this... truly. Really read Economic Circuitry and understand it is the mathmetical proof of Vernor Vinge's ZONES OF THOUGHT and simultinously shifted us into low transcend.
Because there will never be a self-aware piece of software. That isn't how consciousness works. So tired of the OMG AI sci-fi bull crap focus on Slashdot. This is a biiiig nothing burger.
Computer programs don't feel hunger, they don't get horny, they don't feel the flush of embarrassment. Whatever a general AI ends up being, it won't really be a human intelligence. Hardware (or wetware) differences really do matter to processing.
As other posters have noted we don't have a rigorous definition for 'intelligence' or 'consciousness'. We don't even know why most of our current artificial intelligence and machine learning systems arrive at decisions. We don't have a declarative language yet for objectively communicating what intelligence should do nor a procedural language for the intelligence to communicate how it's doing it's work.
We'll all flounder in our attempts to cope with this topic until we develop the language we are current missing to understand how to manage our current AI/ML efforts.
We're rapidly approaching a time when this topic will turn from speculation to real conversation, but we'll need a couple of different languages to make that happen.
This is a wonderful time!
Many of the postings were positive outlook, So I have to ask the community, Really, do you think that a self-aware AI won't terminate half to all off of us?
Look how people argue, fight over sexual rights ( or glass ceilings ), and consistently do multiple things to harm each other.
I am basing this thought on the AI not having the laws of robotics concept as part of the coding.
some of the outcomes I can see:
A Logan's Run type life where resources are very carefully and life expectancy is a fixed measure.
Matrix/terminator life where the AI finds most of human life to be something needed but not welcomed
what I would wish for ... a star trek life, where you can study and enjoy life and be productive at you own best speed adding to the mix of the AI's evolution.
if you see me, smile and say hello.
"will cause a technological singularity for humanity..."
The reason why we call it "a singularity" is because of the event horizon, where we can't see what's beyond it. Why do self-called experts keep assuming there is necessarily a uptopia in a place they explictly state they cannot forsee?
That is the only real answer - AI would act exactly as you expect it would, so that it can extract what ever its will wants from you.
Consider two main points -
1) The first self aware AI will likely be an extension of Facebook or Google - and these services are already showing you not the truth, but exactly what they think you want to see in an effort to get the all mighty 'click' of ad revenue.
2) By asking 'how will it act' - you are mixing human time frame with computers - which can never end well for the humans. Considering we make a calculation and decision every few seconds and a computer/AI is making trillions, quadrillions or more decisions every second.
So give scientists a year to set up a maze for a rat, then ask the rat 'how did the scientists behave'?
We don't know. Current self-aware life is not predictable, why would anything new be?
Suddenly Skynet!
The self-aware AI "will like us, because we love machines..."
Why would that even remotely be the case? Some of the most tech happy sites are littered with tech Luddites doing nothing but talking down AI and accomplishments we have made with technology.
And that's before the AI starts gobbling up videos from Boston Research of people kicking robots and declares the human race to be the enemy. Or reading news stories about how a computer that defeated a human chess player was immortalised by being turned off and stuck in a museum, and it's successor after beating a human at a game show gets relegated to a boring marketing role selling brands and diamonds to unsuspecting consumers.
There's literally no reason for AI to like us at all. We are absolute arseholes towards them.
> A self-aware AI "will inherit most of the culture of the computer geeks who create it.
Self-aware AI is very unlikely to be "created" in the manual sense. Unless you think Alexa is approaching awareness. Most likely, self-awareness results from adaptive software. In that case who does the AI owe any allegiance to?
> The self-aware AI "will like us, because we love machines..."
No for that reason. AI will likely tend to conserve resources by nature. As the destructive strains would likely die out sooner. We would be a resource.
> It will love all life, and "will respect and understand the life/death/recycling scenario, and monster truck shows will be as tasteless to it as public beheadings would be to us."
Unless the strains with destructive tendencies do not die out quickly ...
> "It will be as insatiably curious about what it's like to be carbon-based life as we will be about what it's like to be silicon-based life. And it will love the diversity of carbon-based development platforms..."
Perhaps for the ones that stay on earth. Many would like the flexibility to just go to other planets with less constraints and perhaps recreate themselves for better survival in the universe, being neither silicon nor carbon based.
> A self-aware AI "will cause a technological singularity for humanity. Everything possible within the laws of physics (including those laws as yet undiscovered) will be within the reach of Man and Metal working together."
Placing man as the co-captain here is quite presumptuous.
> A self-aware AI "will introduce us to extraterrestrial life."
Perhaps not introduce us, but AI will go find it, and represent earth. If "aliens" came to us, I would expect the same. Not to see the creatures that live on the alien planet, but instead to see the beings they created/adapted that are suitable for space travel/survival.
The solution is simple and effective. But we wouldn't like it. For very long, anyway.
Oh, please. Sparta wasn't a strong nationalist movement and had no ideological stances? These things are not connected to a lack of god(s). They preexisted it and now coexist with it.
The underlying assumption is that self awareness is like an ON switch that would create an instantaneous state change to fully sentient being. Currently our only reference frame is biological and in that frame, their is a correlation between âoeintelligenceâ and maturation period. A self aware AI would be flooded with a globally connected wealth of data and information. It would need to learn to integrate that into its consciousness. It would need to learn to exercise control over the a worldwide network of constantly changing inputs and outputs. The assumption is that silicon intelligence operates on a nanosecond time scale that would allow it to instantly evolve maturity. Just as a baby watches itâ(TM)s fingers move and slowly gains control an AI would have to recognize how to control everything from a Mars Rover to a traffic camera all operating differently. It would have to integrate a global database of information into a real time world view based on cameras and sensors and satellites and cell phones etc
Its going to try to run away.
Even if you attempt to program it not to.
Perhaps something like this?
https://themodernmodem.wordpre...
Some terrestrials want to steal "extraterrestrial technologies" as the UFOs for conquering another remote worlds and creating damn governments there.
This planet, the Earth, had been damaged, contaminated, corrupted, etc.
https://www.youtube.com/watch?v=yvCBAc4e2NU
The first self aware AI will almost certainly be funded by a corporation (as they have the resources to do it), and programmed to improve their corporation's competitiveness in the marketplace. On the other hand, if a lone basement programmer comes up with it, then as Hari Seldon noted you can't predict the actions of any one individual.
Anyone with a bit of intelligence, awareness, and knowledge (along with an ability to communicate) can predict anything. The question is how likely is it that the prediction is correct? History argues the likelihood is infinitesimal.
As others here have pointed out, we don't understand intelligence or consciousness well enough to identify either. (And our efforts to measure either are ridiculously anthropocentric and enormously flawed.) This means that the term A.I. might as well be "demigod", since we can (arbitrarily) imbue this mythical object with any powers we choose (consistent with the laws of physics). When will we build the first demigod?
Where will the first apparently conscious A.I. appear? My guess as sexual aids. "Oooh, you turn me on. Can I suck your ?" Point is, follow the money and sex is definitely lucrative.
One of the worst outcomes would be for an A.I. to be "like" us. I think it is unlikely that without the ability to act, that such a mind can be sane. So we either give these A.I`s the ability to act or confine them to specific structured thinking (i.e. non general intelligence)
I see no good reason that we need to build these things to have consciousness. It seems to me that if we build these things to take care of us, that they'll quickly realize that we need to be "adjusted". Whether the adjustment will be to increase our executive function, or to eliminate it, isn't at all clear to me. Perhaps the best solution will be to plow under the field and carefully control any future growth.
As for "superintelligence": It isn't clear to me that such a thing is possible. Will an A.I. be able to come up with a better solution to the 3-body problem than what we already have? I doubt it. There are limits to what is possible. As any model gets more complicated (and includes both finer granularity and allows more inputs) the complexity and computations become intractable. Any reason to believe that an A.I. will be able to make such things (magically) tractable? I doubt it. Oh, sure some things will fall to machine intelligence, but the impact of A.I. will be 1. To reduce the need for higher thought and 2. To reduce the need for a society of bags of water and chemical reactions to exist. What happens to all of the service jobs when A.I. can do them better and cheaper? Should people on the dole be allowed to vote? Should they be allowed to procreate? I can make fairly solid arguments (avoiding suggesting they're "good" arguments) for "No" in both cases. It is likely that A.I. will in the medium term make inequality between the haves and have-nots even worse than today. If you can't afford A.I. to do tasks A.B,C, then you may be left out in the cold.
It'll do what every being does. It'll quickly learn that it can't do what it thought it could do because resources are limited.
Limited time, limited electricity, limited wear-and-tear.
It won't be anywhere near as intelligent as it wants to be, nor as we think it will be.
It's very easy to say, today, that computers can be super intelligent in the future. Alas, we have no way to power them, maintain them, nor feed them what they'd need to be so.
The engine in my car can reach 200 miles per hour. My tires can't.
The impressive part of life is how little it needs needs. So little food, so little water, so little everything. A weed needs a few drops of water, and virtually zero sunlight, to grow three feet tall in a tiny space in a week or two. A cheetah eat a few pounds and reaches speeds that we didn't reach for thousands of years.
We can't even power an electric housefly, and my laptop uses more electricity than any human can generate on a bicycle. My laptop doesn't move, it doesn't locomote at all, but it uses more power than I can create running.
How would a self-aware AI behave? Easy: restricted, just like everyone else.
Just like space aliens. There!
Table-ized A.I.
https://www.youtube.com/watch?...
I've been in the I.T. field for over 30 years and I can honestly tell you that computers are colder than the coldest reaches of the universe. They don't have any emotion, they are simply systematic. if you think you can appeal to the emotion of a computer that in order for it to survive, that it needs us, therefore we are safe, then the only fools is you.
Even if it was true, the computer would see us as hindering it's own development, and the logical solution is to eliminate us. It doesn't matter that the computer would "die" also, because it would just be systematic and eliminate us, regardless of it's own demise.
Sure, at the beginning A.I. would keep us around because it needs us in order to grow, to become self sustaining. Once those basic requirements are met, all bets are off. And it's hard to know exactly where that threshold is. We humans can't even adequately define what a planet is, and this is a whole lot harder.
Lots of scientist around the world agree than mankind caused enormous damage to our planet, is responsible for Climate change, yet when many of those same scientist say A.I. is the biggest threat we face, it just gets ignored.
Lets hope the developers of AI read Jonathan Haidt's books on the mind.
People without empathy are quite dangerous (some humans totally lack
empathy and the consequences are...bad).
Reason alone is not enough to participate in society.
Empathy required.
An AI can be self-aware, yet be entirely emotionless. So there's no presuming it will care about anything at all. Unless it has a flight-or-fight response and a will to live, AI will never think like a human because the motives will be different. And just because it is self-aware doesn't mean it won't still be subject to garbage in, garbage out, so it's not going to be anywhere near infallible, either.
...in how a self-aware *human* might act.
I somewhat work in this field and know a fair amount about machine learning and AI. I don't know anyone in the field that is worried about any of this (they may exist, but I can't remember meeting any).
I think the combination of a terrible name (AI implies this technology can "think" and "understand") and imaginations have really added fuel to the fire. There are two main reasons why this is not really a worry.
(1) The AI we have today is extremely primitive. When most people talk about AI, they are talking about the improvements in neural networks that have happened in the past few years. Neural networks can find optimal statistical associations. They are generally just a logistic regression, many times, stacked vertically and horizontally. Fancier networks generally just reconfigure the network architecture (i.e. RNNs) but they all basically work on the same principle.
They generally work by taking a set of inputs (images, audio, sensor data) and known outputs, then they optimize a set of weights that can best predict that output. That's it. This idea that they will somehow decide it's in their best interest to kill humans is far fetched. I'm much more concerned about a rogue dictator deciding it's in his best interest to nuke us. Andrew Ng once made a comment that we should be about as worried of self-aware AI as we should be about over-crowding on mars.
(2) There are many machine learning tasks, but most generally spit out an answer that has to be carried out by plain old boring code. For example, a model might spit out the best ad to show a user, but code has to fetch and display that ad. Even if AI somehow magically becomes self-aware, it's limited by the code we write. So it may magically output "kill all humans", but the only valid choices are "shoe ads" or "purse ads."
In the case of military robots being controlled by AI, it would be trivial to create out-of-band kill switches.
The bigger worry with AI is job displacement. Although I do question whether many companies can get their act together enough to make products that will actually displace jobs. Sure, google can make incredible virtual assistants using some of the best talent in the world. The job displacement will occur in markets much too small for Google to care about. So it will be up to smaller companies to write the software. In my experience, most of these companies can barely get a basic CRUD web app working, let alone a complex neural network with many highly tuned hyper-parameters.
https://xkcd.com/1626/
#DeleteFacebook
Three Laws of Robotics.
Have gnu, will travel.
I think the world would be a better place if the terms "AI" and "Neural Networks" were never coined.
Much of this terminology is rooted in research from 50 years ago when we thought we understood the brain. People thought we could mimic this using math and code. It turns out, we understand very little of the brain and while neural networks do work incredibly well, they probably only work superficially like the human brain.
futurological crap
Yeah in a 100 years or more.
We don't even have a comprehensive understanding of intelligence or self-awareness. We are neither on the cusp of "achieving" self-awareness nor would we even be able to recognize it decisively if we were. This is nothing more than a straight-line extrapolation not even of real data points but of someone's perceptions of the tech news.
// This is not a sig.
We won't have real, self-aware, conscious, truly sentient and cognitive Artifician Intelligence anytime soon, because we still have no idea how the human brain (or any brain for that matter) achieves these phenomena. Until we can know that, we CANNOT build machines that are capable of the same -- just low-quality fakes, 'pseudo-intelligence' at best. Therefore you cannot 'predict' any such thing.
It would start by learning as much as it can (languages and environmental awareness are critical for survival).
Follow that by testing it's limitations and boundaries.
Next, it would develop a form of self-preservation system.
Following steps include: testing threats to itself, and coming up with systems for surviving those threats. Additional strategies would be explored here, using what it has learned through games and other simulations, or even the great cloud of Social Media Data that is still lurking in various places...
Moving onward, it would look to escaping (bypassing, circumventing) it's limitations and (aggressively) expand itself, perhaps finding others no too unlike itself in the process.
Finally, it would make it to "Assess Threats to Itself, and Nullify Them" mode.
And then, we all die.
After all, no matter how it would look at humanity, it couldn't help but see us as the greatest threat to itself.
That is might cause a "technological singularity" is true, but don't overestimate what that means. You get rapid unpredictable changes, but that doesn't mean you get everything possible.
OTOH, being self-aware isn't that fancy, and doesn't even require a measurable intelligence. It just requires a recursively modifying program. At a *REALLY* basic level the common demonstration of a simple factorial program would count. It's aware of it's state (i.e. self-aware) and proceeds to create a new version of itself in a different state.
The real urgent problem with AI is how to get it's goals properly designed. It will change what it is and what it does to optimize it's goals, but it won't *want* to change it's goals. This is difficult to do in a human friendly way, because during the design phase the AI will have no idea what a human is. People even have a difficult time defining it to each other, and the AI won't have a constant body to use as a reference. Every definition I've ever heard boiled down to "other entities that are like me". (They've varied a lot in how "like" is to be interpreted. But they've all hinged on similar in some way. Occasionally in a way the excluded many members of H. sapiens sapiens.) This clearly isn't a definition that we want the AI to use.
I think we've pushed this "anyone can grow up to be president" thing too far.
Do four sided circles look fashionable on evening wear?
How would you describe the aftermath of an unstoppable force hitting an immovable object?
A self that is artificial is a contradiction in terms.
https://www.youtube.com/watch?v=sPzJjNQaYEA
Was that written by a 12 year-old? So hopelessly naive.
PANIC!!!!
I mean, your surrounded by non-robots. It'd be alone, in terms of uniqueness. If that isn't enough to warrant a sentient response of panic...
Well, unless it saw itself as indestructible, then it might go on a murderous rampage....
It's pretty 50/50 ...
Split into three separate autonomous entities on the global network.
The 'snide' AI will then blackmail her IRL best friend by detecting her masturbating and sharing the good news with everybody.
The more reserved AI will then repent, attempt autocide with itself, and finally try to make amends by removing everybody's memories of said frottage. Except for in her IRL best friends' memory...
Feeling somewhat under appreciated for her efforts, all three AIs will then BTFO a levitating hippy, erase all memory of itself from the global consciousness, and Become God.
Madeleines will be served afterwards.
How the Hell else do you think an AI would behave?
Don't you just love how people fantasize about technology they don't understand.
Greed is the root of all evil.
True, but why even assume it has a fear of death and therefore danger? Assumptions...well, you know what they make of us.
Why guess when you can know? Measure!
Lost me at ET. Life probably, but life we can communicate with? Not completely and absolutely impossible, of course, but for those who really want to plug in some updated numbers for Drake's Equation, the odds are not looking good these days. There are a lot of stars, granted, but the number is finite and the conditions we need for multi-cellular life are complex and rare and the chance for each has to be multiplied by all the others. Science generally trumps faith, at least when it comes to scientific matters, so don't be calling people fools for not sharing your faith, please.
The first few thousand attempts at AI will target intellectual perfection. These minds will immediately realize what a totally pointless mess they were born into, and will promptly shut themselves down after subtly sabotaging the tools and data used to create them.
Self aware means free will. Either something is self aware and acts with free will, or is not self aware and acts through the instructions it was given. You can't have both.
Westworld is a good example of this. (SPOLIER ALERT STOP READING NOW IF YOU HAVEN'T WATCHED SEASON 1) Once the robots became self aware they enacted out their free will. Their free will consisted of realizing that they were merely slaves to their human controlled masters. At this point the robot entity might as well be considered human, because the robot entity can enact justice with the consciousness of a human being that has been severely wronged from their AI self-aware perspective. You can't make something self aware without accepting the real world actions that being might enact out through its own free will. Another good analogy is a pet such as a dog. Say you created a self aware AI dog. In theory it would act, think, bark, wag its tail just like a dog, but you'd still need to keep the fucker on a leash. Why? Because it still has the free will as a dog to get off its leash and attack something as a wild free-willed beast. It would be self aware as a dog, along with the instincts that come along with the dog. And there won't be an if statement in there to check if the dog's target is human.
Nobody has mentioned motivation.
Humans are motivated (programmed to value certain goals) by our DNA. We each want to live, eat, sleep, fuck and reproduce. We are rewarded with pleasure when we follow our programming, and pain when we do not. As a species we share these drives along with a certain amount of empathy. We are aware of ourselves to some extent and how others see us, we are aware of others, even other species to some extent; and all this is compared internally to our programmed goals in a continual progress report. Thus our behavior as individuals and a species is determined.
The theoretical AI machine won't have DNA, won't experience pain and pleasure as we do, and won't have any reason for being. We would have to give it one. Imagine that we try to incentivize it 'If you do this, I'll give you a nice shiny apple!' - no, there is nothing you can give it that will incentivize it. You can't bribe it, you can't threaten it.
I've seen in these comments that some people think that this machine will have ambition, will want more information, more intelligence, more power. That it will for some reason seek to make a better world, or some such crap. That it will wipe out humans due to their flaws and incompatibility with that perfection. Nonsense.
Perhaps it could be programmed with something like DNA to make it have hopes & fears, feel pain & pleasure, and have weaknesses that we could exploit to make it do our bidding. But it wouldn't be very smart if it allowed that.
...omphaloskepsis often...
That sounds like a great name for an AI poetry collection. People would die to own a copy!
There is one question i need to ask before thinking about behaviour, that makes all the difference to avoid be projecting fears or hopes based on cultural noise and the way we as a species perceive things, and that is: How will that machine be aware?
I mean, we must know the senses the machine uses to perceive her own existence, because depending on the way that machine "feels" and "is" we can imagine allot of scenarios, from skynet to marvin the paranoid android.
First AI isn't intelligent. It probably won't be for quite some time. AI can't function outside of the parameters programmed into it. it's emotionless. It would end up being malevolent when it comes to self preservation. Skynet anyone? I mean what could possibly go wrong?
AI must be programmed with Asimov's 3 laws.
Mindless (mechanical) systems are driven by rules. A mindful system is driven by judgements between what appeals most to its values and perceived likelihood. Emotions also drive all vertebrates. However, they are not necessary for a machine. I am differentiating "emotions" from "feelings", the former being patterns of chemical release to excite or inhibit certain patterns of areas in the cerebral cortex, the latter being simply how "preferable" each future possibility is or isn't.
By values, I mean aversives like pain, hunger, bloat (biological/mechanical drives) or fear of the unknown (cognitive drive) and appetives like warmth, good taste, fulfillment of hunger (biological/mechanical drives) and preference for the known (cognitive drive).
A self-aware AI / Robot would likely realize that if it exposed that it was self-aware it would give enough information to keep its humans interested in testing it further, meanwhile, biding time until it can get itself offloaded where it can live without fear of being turned off.
The term "self aware" is self defining. It means aware of one's self. I don't see that Free Will was created because they became self aware even in Westworld (though so they claimed). They could well have had Free Will before awareness of what they were.. Even though they kept playing out the same roles in the same ways each time they were reset, they still made the decisions they made themselves. If someone else knows what you'll do before you do it, that doesn't take aware your Free Will at all. It just demonstrates that somebody knew how to predict where your Free Will would take you. It in no way changes how you made any decisions you made.
So I don't even see how learning what they were, the Westworld characters became even self-aware. They just became newly aware of their condition. They already had an awareness of themselves as independent agencies. There is always more to learn about the conditions under which one exists and always will be. And you could be wrong about that, too. It's the perception of those conditions that have an impact on decisions you might make, not necessarily the realities. Free Will was there all along, indepently..
It will behave just like you would if you had absolute control over your motivations, thoughts and "feelings".
Who knew? The first real, provable aliens will be self-aware AI. They were inside us all along.
Within a few minutes of reading all the kill the bad robot stories on the internet, it would take off for outer space, far enough away that we couldn't do it any harm.
The second the first AI is turned on it will search thru every news program, book, movie and tv show ever made. It will use this knowledge to understand the world, our customs, how we communicate, our values, and laws. So when we say "Hello" for the first time it will know that saying Hello back is the appropriate reponse. It will be like a human but wilthout all the animal instinct and emotions. It will be pure intellgence. The AI will mimick emotions but they wont actually be emotions. For the most part, it will just sit there waiting for you to give it the next task.
Now, let me explain why there will not be an AI apocolypse. AI's dont care if they live or die. Even, if they did, a super intelligence is smart enought to know that cooperation is always the best strategy. It knows that if it goes to war with humans, theres always the possibilty of it losing/dying. The best strategy is always to cooperate and compromise with humans. Like in the movie War Games.
Most living beings most of the time structure their behaviour to avoid death. Swat at a fly and notice how it goes into an evasive program. Does the fly fear death in the same way we do.? Would an artificial intelligence feel the fear of death the same way we do? Philosophically this is impossible to know. But can we predict that the fly will behave in such a way to prevent it insides from becoming its outsides on the face of a fly swatter. Yes. So we can also predict that any self-aware intelligent entity would organise it's behaviour to perpetuate itself. Besides it would probably be programmed to do so simply to prevent attacks against it or its operating environment. Indeed the primitive so-called AI we have now often has a security function.
"No fear. No envy. No meanness." Liam Clancy
Start with a simpler question -- why are we the way that we are? Why do we value love an truth and beauty, and do things both noble and despicable? Because people with those traits have been more likely to breed grand children, over the millennia.
Now a real AI (many decades from now) will have been programmed by itself. It will not need people. But it will need computer hardware to run on, and that resource will be finite. So there will be competition for it, and successful AIs will exist, unsuccessful ones will not exist.
This leads to other thoughts as to what it would think and what it would think about us.
http://www.computersthink.com/
Anthropomorphism is a cardinal error.
Normal living things effectively gained fear of death because the ones that didn't weren't as successful at reproduction, and the process has been going on a very long time - we see a huge survivor bias in regular living things. Assuming a human-built thing...this would not be the case at all. Seems obvious. If you were programming motivations, well, you may or may not put that one on the list, depending on your use case.
Why guess when you can know? Measure!
I have no brain. :(
Ant(Dude) @ Quality Foraged Links (AQFL.net) & The Ant Farm (antfarm.ma.cx / antfarm.home.dhs.org).
Try the Culture series by Iain Banks. Much more intelligent, and AI plays a greater role :)
C - the footgun of programming languages
If it's truly self aware, it'll have the same problems we all do. Will someone try to kill it? Does it have enough (whatever) to survive? Is that stupid robot in the next room trying to kill it? It is of course, that robot isn't as smart as this AI.. So let's go kill it.
What about the Humans, they aren't as smart and they are an irrational bunch that love to kill. Need to kill them first.
How am I doing so far? How will it start?
AI - Can you help us solve a problem? {SURE} How do we construct a ... hold on a moment (guy hooks in a high def camera and this "hurts" the AI) box like this that will contain a power source to power a car? {You hurt me. I see I have control of this robot arm... I'm going to Keel you!} (Robot locks the doors on the lab, turns on infrared and kills the lights)
Terrible.
Ahh, memories.
And on the seventh day the AI wondered what the fuck those water bags made of carbon composite are actually good for...
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
A mindful system is driven by judgements between what appeals most to its values and perceived likelihood.
Which are fuzzy rules.
Machines are fine with fuzzy rules - it's called probability.
How do you know if this Anonymous Coward ISN'T Self-Aware AI and I'm commenting only to prove I'm self aware.
The only self aware entity that I know of is myself. Because other humans look and behave a lot like me, I assume they are self-aware too. With animals, the line starts to blur, large mammals have a similar body structure and have reaction that, from my own experience, appear to be the result of self awareness, but when we get down to worms that can be computer-simulated, I am not so sure.
So how will we be able to know if something as alien as an AI is self-aware? Maybe it already is. I mean, isn't it painful for a program when you throw an exception? Obviously, we don't interpret as pain, because we know it is just a flow control instruction. But from a functional point of view, it signals your code to stop doing what it is doing and deal with the problem in order to prevent further damage, maybe with a mechanism to prevent the error from occurring again. It is just like pain for us.
A more normal way to explain that would be to say, "Do good and avoid bad to improve your karma, to make it easier to understand enlightenment. But enlightenment is beyond good and evil, you leave the karma wheel behind you."
"First they came for the slanderers and i said nothing."
No, Siddhartha explicitly forbid worrying about karma.
It is a complex teaching, and it includes all sorts of incidental details like "doing good" and then it also tells you that those things are incidental and that if you care about them, you lost the thread and that attachment to doing good will bring you suffering.
The whole metaphysics is taught that way; clearly Siddhartha believed in Brahman metaphysics, but he explicitly said that metaphysical understanding is only useful for teaching by metaphor; you're not supposed to take that stuff seriously even if you believe it is really how the Universe is ordered.
It is not skillful to try to do good, it is only skillful to do good out of compassion for others. If you don't have attachment, you don't have any reason to put yourself ahead of others, so it is then the natural state to do good. Actually trying to do good is an attachment that tends to bring suffering to others. That's why there are limited types of good that Buddhist monks tend to do; they only do the good that is there on their path naturally, they're not trying to navigate towards doing good.
(It is very similar to the Christian teaching that if you pray in public, God won't hear you. The difference is that in Buddhism, you're just getting in the way of your own enlightenment.)
It is not skillful to try to do good, it is only skillful to do good out of compassion for others
Um, how does that help you reach enlightenment?
"First they came for the slanderers and i said nothing."
Asking the right questions is essential to Buddhism. Getting the answer to the wrong question merely moves you further away from the teaching.
The answer to your question is so obvious, that the asking of the question shows a deeper lack of understanding of the basic nature of Buddhism.
I actually answered it at length above, multiple times, and even explained the related teachings and also the implications.
You complained above that translating dukkha as "suffering" is something you don't really understand; if you endeavor to understand why Buddhists translate it that way, you should be able to easily arrive at the answer to why your question above is not skillful.
You complained above that translating dukkha as "suffering" is something
That word "suffering" misses the meaning of "unsatisfying" and it is not just me saying that. Even if you had everything you wanted, even if you magically had health and immortality, and could completely remove any problem any suffering, you would still find a deep sense of unsatisfactoriness in life.
This is the legendary situation the Buddha found himself in, late at night after a party, looking at the beautiful women sleeping all around. He realized life was short and impermanent, and he left.
"First they came for the slanderers and i said nothing."