Visitors of this forum might find the new site 3 Laws Unsafe (http://www.asimovlaws.com) interesting:

In anticipation of 20th Century Fox's July 16th release of I, Robot, the Singularity Institute announces “3 Laws Unsafe” (www.asimovlaws.com). “3 Laws Unsafe” explores the problems presented by Isaac Asimov’s Three Laws of Robotics, the principles intended for ensuring that robots help, but never harm, humans. The Three Laws are widely known and are often taken seriously as reasonable solutions for guiding future AI. But are they truly reasonable? “3 Laws Unsafe” addresses this question.

The site contains various interesting articles (http://www.asimovlaws.com/articles) that expand on the problems in using an approach based on Asimov's laws to ensure a robot or AI's ethical behavior. For example, the meaning of these laws could be twisted in various ways, as in Asimov's stories; it could turn out that in retrospect, we wouldn't want robots to follow these exact laws at all. Then, there are the ethical problems in constraining a mind in this way. I think the arguments to dismiss the three laws as a model of AI ethics are convincing.

Are the three laws actually taken seriously outside SF?

I think so.

Sometimes, yes. There are also weird variants (http://www.angelfire.com/rnb/fairhaven/ethical-laws.html) that suffer from the same problems.

One of the most seemingly problems is that artifical intelligence may use the three laws to their own advantage.

I dont think any government has made these laws legal yet, so until then lets think about how to create these ...

I dont think any government has made these laws legal yet, so until then lets think about how to create these ...

Actually it has been said by scientists that 2035 is a practical year for I, Robot to be set in. Because of everyday advancements in the world of technology, its safe to assume that we could buy our own personal household "NS-5" by 2035.

Just thirty years ago the flat screen computers we use nowadays and mainframes that could be carried by hand were a fantasy for those computer junkies back then.

A big problem is in implementing the three laws, how will AI in the future work and how will it be controlled into safe non-revolting servants.

There several approaches I see based off how the AI works:

Analog implementation of the 3 laws: Digital computers don’t have much of a hope of functioning like us, digital logic processing does not work in the varying world of, us human with are analog processing might be a better model, so AI running of Anolog processors (or composite analog digital processor) might be the future. Anolog system are difficult to program, their programming is soft and variable, the 3 laws as directive could not work, instead like human directive would most likely be implement emotionally: here is what the 3 laws would look like.

1. Do not kill people, if possible prevent them form dieing. This would be implement with extreme displeasure in killing or letting people die, a ultra-high sense of empathy.
2. Always follow human orders unless they violate rule 1: the only source of pleasure is from doing what people say, like a little orgasm from following human orders.
3. Do not kill ones self unless it violate 1 or 2: robots that can feel pain.

Emotions make us eat, sleep, fuck, live. In humans though it is possible to override our Emotions, people commit suicide for example sometime for an ideals like pride and honor over a emotion lie depression. A robot would need to have far more persuasive emotions then a human as well as weaker will power to over come.

Autistic control: who says AI has to think like people, at present computers and robots do not revolt, they are incapable of such thoughts as all they do is crunch numbers. Ai could be made of this extension, a very intelligent set of robots could do all the manual task as human does, say you could have a service store run by robots, they stock and check out goods, sweep the floor, ect. And none of this requires the mental capacity for abstract thoughts, thought such as killing off people and world domination. Such autistic robots who are very skilled and mentally inept would be ideal for the military as killing people is what they would do yet they would be to limited in thought to consider going AWAL. Such robots might never seem human in behavior, nor would they be able to do everything that a human can do.

Asimov himself wrote a great deal about this very issue, a bunch of his later stories revolve around the different problems the three laws could cause.

With our limited knowledge of the workings of hypothetical robot minds, it seems there are two main choices;
robots can either be idiot savant, autistic, emotionally deficient and constrained to obey the Three Laws (or four or five, or 'n', however many it takes) without question;

or they can have mentalities more closely resembling the average human mind, and obey the 'n' Laws because of emotional and instinctive imperatives, rewarded by pleasure when obedient, punished by pain when in contravention...

but this 'analog' strategy would allow a strong-willed robot to overcome its conditioning and disobey the 'n' Laws supposedly governing its behaviour.

For example; fear of snakes and of heights are both supposedly innate fears/behaviours in humans; both can be overcome. Such imperatives as sex, eating, even breathing can be overridden by a strong-willed human; Such an act of will would be open to a sufficiently human-like robot.

In fact it would perhaps be an infringement of that human-like robot's rights if it were constrained to obey any of the hypothetical 'n' laws of Robotics, whatever they are eventually conceived to be.

SF worldbuilding at
www.orionsarm.com

eburacum45, I disagree with your two options, unless you meant to say "if the robot is to be based on the three laws, then we have two options". This approach (http://www.singinst.org/friendly) falls in neither of your categories.

As I understand it, a robot can have (or at least fully understand) emotions and be non-autistic (in the sense of being able to make intuitive sense of minds) without being human-like, or being motivated only by pleasure and pain, and so on.

In any case, imposing laws on a robot that it doesn't want to obey doesn't work. But as far as I know, there's no reason (in principle) why you couldn't build a robot that doesn't want humans to be harmed, and that can understand what the "spirit" of the laws is, so that it could decline to follow them if that's not what we would want if we had really thought about it.

and how do you impose the a desire not to hurt humens? are you saying we could program them in as higher ideals, as morality?

Yes; I did mean to limit my case to discussions of the Three, or 'N' Laws; I personally do not think any kind of law can be programmed into a fully sentient being.

I have always been an admirer of Yudkowsky's attempts to produce a framework to begin the development of friendly AI; but his approach is not likely to be the only route towards artificial sentience, and so I don't expect that friendly AI will be the only result of the process of emergent intelligence.

It would be nice to think that all AI will be as friendly as Yudkowsky expects; but once these entities are given the ability to delf design and self evolve, they will develop in any way necessary to best face the various challenges of the unforgiving Universe; we can only hope that we are important to their plans.

SF worldbuilding at
www.orionsarm.com

The three laws are nonsense. What is considered harmful in one culture is a necessity in another, to give a robot the power to make a moral decision is dangerous. The only law I consider safe for robots is that a robot may not do anything that its owner or owners have not explicitly ordered it to do. It is the humans in control that need to follow laws.

Robots have already killed humans and will continue to do so.

It would be nice to think that all AI will be as friendly as Yudkowsky expects; but once these entities are given the ability to delf design and self evolve, they will develop in any way necessary to best face the various challenges of the unforgiving Universe; we can only hope that we are important to their plans.

Yudkowsky certainly doesn't expect all AI to be friendly (not sure if that's what you're saying). On the contrary: I'm pretty sure his view is that, unless you know exactly what you're doing, building an artificial general intelligence is a significant existential risk (http://www.nickbostrom.com/existential/risks.html). Basing your design on Asimov's laws is just one way of not knowing exactly what you're doing.

As for self-design: in theory, if an AI starts out not wanting to harm us, then it will be able to see that, by changing itself into something that doesn't mind harming us, it will be indirectly harming us, and that this will be undesirable. For this to be safe, the AI needs to already be sufficiently smart/wise when given the ability to redesign itself, I guess.

Robots have already killed humans

Really? Where/when? (Just curious.)

Dose a bullet know any laws ? hence ...

As for self-design: in theory, if an AI starts out not wanting to harm us, then it will be able to see that, by changing itself into something that doesn't mind harming us, it will be indirectly harming us, and that this will be undesirable.
The AI might discover a Zeroth Law that Humans are bad for the universe and need to be removed.

For this to be safe, the AI needs to already be sufficiently smart/wise when given the ability to redesign itself, I guess.

Just as humans will need to be smart/wise when we gain the ability to redesign ourselves; this is likely to happen fairly soon.

Robot Man Killers
http://www.cdc.gov/mmwr/preview/mmwrhtml/00000505.htm

Robot safety....
http://www.osh.net/articles/archive/osh_basics_2002_may24.htm
http://tis.eh.doe.gov/docs/osh_tr/ch1d.html

The AI might discover a Zeroth Law that Humans are bad for the universe and need to be removed.

"Bad for the universe" in what sense? Would the AI necessarily care?

Blindman,

Thanks for the links.

As for self-design: in theory, if an AI starts out not wanting to harm us, then it will be able to see that, by changing itself into something that doesn't mind harming us, it will be indirectly harming us, and that this will be undesirable. For this to be safe, the AI needs to already be sufficiently smart/wise when given the ability to redesign itself, I guess.


Why AI should care? Do we care about other species?- we don’t give a crap most of the time . I mean "friendly" AI probably wouldn't harm human just for sake of inflicting pain -it could be just a side effect of it's activity . - Same way humans caused immense damage to wild life as a by product of their life activity.

It could be direct competition over resources/land . Humans are dangerous as they possess necessary power to destroy even higher forms of organization with ease (nuclear weapons) that’s why AI may decide to eliminate them .- Heck humans destroyed a lot of wild life species just because they caused inconvenience- not even because they posed any kind of threat to humanity as a whole.

I Robot ?

What's the need for AI? I'm with the Fetus' Autistic design. But, AI is going to be designed, whether there's a need for it, I'm certain.

Perhaps we could instill in our creation a love for its creator. I mean, Christians love their God, despite His foibles, so why can't AI love us, simply for creating it? We just can't get any goddamn atheist AI programs....

ROMAN, Do U know matrix, Do U know What it is, Do U know What it will Do to U humans, DO U no Y itill Do that 2 U ?

screw the 3 laws, don't enslave the robot, let it behave in society as a human, and who knows, maybe it'll act and be treated like human after a while! let it vote, run for president, be a dock worker, whatever!

Well, Blindman, those machines are known as "robots" but they are not robots in the sense of this discussion - merely more complex industrial machinery. Robots of that nature will undoubtedly kill by accident in the future. Robots of the sapient type we are discussing here will also undoubtedly kill by accident, for all the safeguards you might put in. This isn't really germane to the discussion.

It was sad, I thought, that the Friendly AI people seem to have misunderstood Asimov:Friendly AI is an attempt to get rid of the concept of "Asimov Laws" (a science-fictional plot device invented in the 1940s) and replace it with a serious discipline.
Friendly AI is an attempt to get rid of the concept of "Asimov Laws" (external programmatic constraints on an AI) and replace it with a solution which works even if the AI has unrestricted access to its own source code. Clearly from the second one they didn't understand Asimov's concept whatsover. Although the three laws were built in, they were not programmed in the conventional sense. The actions of a robot were determined by mathematical laws which determined the interaction between positrons in the "positronic pathways". In one story he hints that removal of the first law would yield "no non-imaginary solutions to the positronic field equations". So the robots are not programmed to not harm humans, they are constrained to do so by the laws of mathematics and physics (in his stories). By claiming that their aims are contrary to the Asimov concepts, they imply that Isaac Asimov would not have wholeheartedly endorsed their program, which I'm pretty sure he would have. It seems to me that Friendly AI aims to produce exactly this kind of safety in purportive robot sapience.

All this is a storm in a teacup - I'm quite convinced (by the arguments of Roger Penrose, see The Emperor's New Mind) that AI will never approach human levels of intelligence, insight and indeterminability. They will do what they are programmed to do, and they will be programmed to avoid harming humans whereever possible - and like all programming, it will be fallible.I Robot ?

I didn't understand what you were replying to, but are you referring to the book (pro-Robot, Humans wrongly don't trust robots) or the movie (anti-Robot, Humans wrongly trust robots)?

I'm quite convinced (by the arguments of Roger Penrose, see The Emperor's New Mind) that AI will never approach human levels of intelligence, insight and indeterminability.
And I am convinced by the arguments of Vernor Vinge that there will be an artificial intelligence which equals humanity...
briefly.

The next day work will start on an upgrade.
Or perhaps the day after- following the party...

Never heard of him, so I googled him and came up with a Salon article, from which I extracted the following:"The singularity" occurs in that moment when computers become intelligent enough to upgrade themselves. Self-programming computers will have, argues Vinge, a learning curve that points straight up.If the first occurs, the second will absolutely occur, and far faster than we would imagine. The doubt raised by Penrose is to whether the first can possibly occur. My view is that computers will never be intelligent enough (as programmed by human beings using conventional one-dimensional mathematics) to be able to "see" in what way they would require upgrading, or to intuit other ways of upgrading and improving their abilities beyond their current ones.

Our ability to "see" not only the evident solutions to problems, but the hidden solutions, the solutions to problems that haven't even arisen and the concept of problems that may never arise, is an attribute that is not and never can be governed by conventional, computable mathematics.

That's right; it seems to be a case of 'either, or'.
Either
human consciousness is not replicable artificially, in which case the singularity is unlikely to occur;
or
human consciousness can be replicated, in which case the limitations of the human mind will decidedly not apply to self-programming, self designing machines.
and the singularity is inevitable.

you all will evolve into RoboSapiens

So the robots are not programmed to not harm humans, they are constrained to do so by the laws of mathematics and physics (in his stories).

I'm not convinced that this difference matters: the issue is still one of imposing external laws on an already existing mind with other goals, vs. designing a mind that is nice to us because it really wants to be.

I think that whatever Asimov actually meant, when talking about applying Asimov's Laws to real robots, people mean programming in a few laws as constraints on behavior. The important thing isn't whether Asimov was right or wrong; it's that this latter approach doesn't work.

As for Penrose: first, I think he's wrong about the human brain needing noncomputable quantum gravity components to see the truth of mathematical statements. Second, even if an artificial intelligence can never really be conscious, or really see the truth of some mathematical statements, I don't see why it couldn't still become smarter than humans in many ways, and still achieve a sort of singularity, and have just as much of an impact. Third, even if Penrose is completely correct, there's no reason why we couldn't use whatever the human brain uses to build an artificial intelligence out of. If biology can build it, then so can we, when in possession of sufficiently advanced technology; and I doubt that we wouldn't still be able to tweak things to make such an AI a lot more intelligent than any human.

I think youd need to give robots a religion,if they harm humans theyll go to hell and be confronted by a robot satan,and be thrown in a river of...er water.

If they serve man well theres a place for them after thier expiry date in silicon heaven with other electronic goods like calculators and washing machines etc.

Sure there will be robot religions; they might worship Humanity for a start, as their creators...

or they could subscribe to any of the major religions, or even minor ones...
as creatures experiencing the majesty of the cosmos, they could develop, or be programmed with, or even adopt voluntarily, a sense of wonder;

when they upgrade their 'positronoic' or otherwise minds until they are much larger and more competent than our own, how could they fail to experience the mystery of the cosmos even more profoundly than we do.

Some religions of the far future here (http://www.orionsarm.com/topics/religion.html)
including this specifically robotic creed (http://www.orionsarm.com/topics/Kja_Observance.html)...

when they upgrade their 'positronoic' or otherwise minds until they are much larger and more competent than our own, how could they fail to experience the mystery of the cosmos even more profoundly than we do.

When their brains are much larger and more competent than ours, the cosmos will probably no longer be mysterious to them.

Besides, religion doesn't automatically follow from a sense of wonder. I think robots won't develop a religion unless you program it in on purpose, and I think that would be a very bad idea.

When their brains are much larger and more competent than ours, the cosmos will probably no longer be mysterious to them.

Besides, religion doesn't automatically follow from a sense of wonder. I think robots won't develop a religion unless you program it in on purpose, and I think that would be a very bad idea.

Well there is plenty of humans who are "programmed" with religion who live "normal" happy lives and dont hurt a fly.

If they can do it,so can the fucking robots IMO,although make it so there is no contradiction,for ex adam and eve is a contradiction of modern belief in evolution and biology,so gotta learn from past mistakes.
You cant grow women from your ribs,if you could id have no ribcage left.

Im pretty certain a robot without a religion that has intelligence will decide its an evolution above us (and rightly so id imagine) and decide it dont need the human race.I mean wed make them slaves to our use,i reckon they would get pretty pissed off at that fast.

In terminator,the t-800's come off an assembly line,cos the cyborgs reproduce,not in mating terms but in machinest terms they can build thier own army.

Lets compare it:
new adult killing machine cyborg built in 1 day

new human adult killing "guy" built in 18 years

duh!? i wonder who will win?

In the beginning reproduction units will be only a few,like cars,but later as they build in numbers the reproduction units increase.

It will be in robots best interest not to harm us, or else ...