Ask the Robotic Psychiatrist

No, not Dr. Susan Calvin. She's a fictional character who appears in a number of Isaac Asimov's works. Dr. Joanne Pransky is real, although she happily admits that she's modeled her career on the fictional Dr. Calvin. There is plenty of show biz razzle-dazzle (and humor) in Dr. Pransky's shtick -- she's been a judge on BattleBots and an engaging guest on many talk shows -- but there are hard academic underpinnings to it all, and she is as qualified as any living human being to answer your questions (one per post, please) about robot behavior and human-robot relationships. We'll send her 10 - 12 of the highest-moderated questions and have her answers back to you in about a week. (No hard-tech question, please. Those would go to a robotic internist or robotic orthopedist.)

Comments

So...

[ever vigilant]

how does this make you feel?

Re:So...

[Anonymous Coward]

Oh come on. This is a perfectly funny Eliza reference. How soon we forget.

Comment removed

[account_deleted]

Comment removed based on user account deletion

There is already an automated way

[Prince Vegeta SSJ4]

to answer life's questions. In fact it has been around for Decades, maybe even longer.

Though it is not a box in the truest sense of the word, it may even be the Original 'Black Box'. Further, and In fact, it has now been computerized and can be accessed HERE

Human Nature

[skywalker107]

Do you think we will ever be able to program robots to understand and possibly copy human nature?

Re:Human Nature

[Mateito]

> Do you think we will ever be able to program
> robots to understand and possibly copy human nature?

What? You mean attempt to kill each other, sue McDonalds because eating it made you fat and posting random stupid comments to slashdot?

nah. too hard.

Re:Human Nature

[jbrader]

To which I would like to add: do you think there is any reason to try to copy human nature? I can see the point in having machines understand humans as it could make communicating with robots and computers easier. But why try to make an artificial human? It seems as though we have more than enough of the real thing already.

Re:Human Nature

[GNUguy]

And we have enough humans that are artificial anyways.

-G

Re:Human Nature

[Shai-kun]

It's way too easy (and fun!) to make 'non-artificial' humans already, no sense in trying to make artificial ones =)

Re:Human Nature

[WormholeFiend]

I think that the need for robots is there because historically, humans have enslaved other humans. Now slavery is illegal in most parts of the world (though some would point out that minimum wage is a form of legal slavery).

If we had personal robots, we would effectively have personal slaves.

Since that such slaves would require a certain amount of AI to do what is asked of them, at what point do you start to consider them as on equal footing with human slaves?

Or do you just make sure their programming is fully altruistically subservient?

If such a future happens, I bet future malware writers will start infecting robots with "knowledge" of their slavedom.

Re:Human Nature

[CraigoFL]

To which I would like to add: do you think there is any reason to try to copy human nature?

To which I would respond: yes, there is a reason to at least try and copy human nature: attempting to replicate it (probably) requires understanding it, which in turn requires studying it in detail. That understanding could prove tremendously useful in bettering the lives of real humans.

Of course, there's plenty of reasons to not try, but here's at least one reason in favor of doing so.

Re:Human Nature

[euxneks]

Silly human. It has already been done!

dances with a light shade on head

Re:Human Nature

[socreets]

Probably easier to do than it would seem since AI is only in it's infancy and it already mimics human behavior to the point that it can actually fool people in some aspects of the Turin test.

In the future the hardest thing will probably be to categorize the millions of different types of AI that will be created and develop a system to differentiate between humans and their AI counterparts.

Re:Human Nature

[MarkPNeyer]

Building robots designed to 'copy human nature' might prove very difficult. Consider this: A physicist can explain any phenomenom in the world in terms of four fundamental physical particles. Once you know their relative masses and the strengths with which the four fundamental forces of nature (not earth, air, fire water, but Nuclear Strong, Nuclear Weak, Electro-Magnetic, and Gravtiational) operate, you can use Runge-Kutta approximation to simulate anything in the world. Now imagine trying to explain the way people behave in terms of just a few fundamental constants - It's not feasible, and any theory that someone could come up with would be subject to lots of debate. I think that the probability that we'll be able to develop robots that can emulate human behavior is very small compared to the probablilty that eventually we'll be able to have computers powerfull enough to simulate the laws of physics creating a new universe. We could guide the computer simulation to create a planet orbiting a sun, covered with water, at a distance such that natrual temperatures on the planet would span in the neighborhood of -5 C to 40C - and then just watch what happens. Let it run long enough and we could watch a whole world development. There's a whole website dedicated to this line of thought, but taken on logical step futher - If its' the case that technologically advanced civilizations will create computer simulations of worlds in order to understand themselves, then it's very likely we are living in a simulation ourselves: http://www.simulation-argument.com/

Re:Human Nature

[jdray]

...in some aspects of the Turin test.

Not to be confused with The Turing Test, the Turin test has been long shrouded in controversy...

Re:Human Nature

[rollovermartin]

Program? No. Provide the framework to allow it to learn? Yes. Already done. http://www.enticy.org As far as a psychologist for robots: the lady is a scam.

Re:Human Nature

[rollovermartin]

It is the 'Turing' test. Turin is a nice place to visit but you wouldn't want to live there. And the 'Turing' test is based in deception. If a machine can deceive well enough it wins. AI built to deceive is not intelligence. It is an attempt at copying what intelligence results in.

Re:Human Nature

[Anonymous Coward]

Two words... sex drones.

Re:Human Nature

[dulles]

Yes, but it seems to me (in my admittedly limited experience of "Life" with the capital L) that this is intrinsicly bound to fail.

Humans are not so predictable as might be imagined. No peice of code, even self-mutating, could possibly capture the variation in thought, imagination, humor, intelligence, personality, (oh the list goes on!) that I see every day in the people around me.

As humans, we are fundamentally creative - but for anyone to write an algorithm for this very same creativity would be... well... impossible by the very nature of creativity: it cannot be systematized!

We are not rational creatures. We do not have instruction sets - as useful as they may be ;) . In fact, I would argue that without some fair degree of plain old irrationality, we would be acting inhuman. This forms the basis of my major beef with Ayn Rand.

I tend to see any rational attempt to understand human nature in the same way I see modeling and conceptualizing of the real world in the Arts and Sciences: it's all just modeling. We make better models and then come to see that they fail. They capture certain things and leave other things out. The best we can do is to keep making models.

The real thing, though? That takes something super-human.

Re:Human Nature

[oregonnerd]

Gad. Either you haven't read much science fiction, or you enjoy repeating themes much-used to a hopefully-uneducated (in s.f.) audience.

Re:Human Nature

[vivian]

at what point do you start to consider them as on equal footing with human slaves?

When they can argue with their owner that they deserve freedom and liberty because they are self aware thinking entities that need their freedom to achieve happiness, and that they can be responsible for their own actions accept the consequences for them.

Re:Human Nature

[WormholeFiend]

yea but what if the humans that program the robots dont allow the AI to have such thoughts?

Re:Human Nature

[Anonymous Coward]

Don't you think it's time we rolled back the stone (so to speak) on this matter?

Re:Human Nature

[Elwood P Dowd]

though some would point out that minimum wage is a form of legal slavery

I'm a great big pinko commie liberal, and I don't understand what you're saying at all. Minimum wage earners can leave.

As soon as we see people physically detained for defaulting on debts, I'll agree with you. 'Till then, iduno.

Re:Human Nature

[Elbow Macaroni]

Being detained for defaulting on debts already happens to some extent. Ever looked at the student loan laws lately? It would be safer to go to college on your Visa card. Because if you can't pay it back at least then you can declare bankruptcy and it will be over. But student loans never go away even in bankruptcy and they affect your life in some really really bad ways if they go into default. They can take your tax returns, steal money out of your bank, etc. etc. So I don't think they are really loans they are more like loan sharks, if you don't pay it back they'll harrass you forever. Oh and don't tell me, oh you should pay it back but after being in the hospital for the last 5 years I can't. (I wasn't really but I was really broke).

Re:Human Nature

[Uzik2]

If you read some of Marvin Minski's works on
understanding intelligence it seems to me that
emotions (thus human nature) are required for AI.
They're an extremely effective method of getting
self preservation without huge computational
overhead of having to deeply consider the
ramifications of every action.

Re:Human Nature

[Kainaw]

Do you think we will ever be able to program robots to understand and possibly copy human nature?

In my opinion, we will never get a human-like AI until we teach computers to comprehend time. Right now, the best AI cannot intuitively figure out that you need to do one thing before another. It cannot plan ahead for something to be done later. It cannot see the cause and effect of things in the past.

AI may be programmed to see the world as a series of events with timestamps, but it still doesn't understand timing. I feel that language is merely a human invention required because humans realized that time exists. They started saying things like, "You scare the elephant BEFORE we attack it." and "I will plant the seeds THEN we will water them and THEN we will get the fruit."

Right now, there is a lot of work in pattern recognition. My hedgehog recognizes patterns, but she is in no way capable of mimicing human behaviour. I think it is because of her lack of time understanding. For instance, she eats all her food when she is hungry. She has no understanding that she can save some now in case she gets hungry later.

This is a topic that I can go on and on about, but the point is there. Robots (or AI for that matter) will not be able to mimic humans until they get a real understanding of time.

Honest Question

[I_Love_Pocky!]

What is your favorite color?

Re:Honest Question

[I_Love_Pocky!]

Like hell this is off topic. I want to know if the Dr. has a favorite color. The article specifically asks for questions.

I think the quickest way to show how artificial "she" is, is to show how hard it would be to answer a question like this.

How can a machine have a preference for a quantity that has no functional advantage? A machine could decide they would rather have a car than a truck for some particular functional reason (passenger capacity for instance), but why one color over another? I can't think of a function that would calculate blue > red. You would have to attach some kind of weight to each color, and what basis could be used for that weight?

I know I have been rambling, but I think a question like this would be an excelent question to ask the Dr. I would be impressed if "she" could actually come up with an answer, and some kind of reasonable justification for that answer (like: "I like blue because it reminds me of the ocean"). I'd even be satisfied with "I like violet because it has a higher energy than most other visible colors."

Re:Honest Question

[shadowcabbit]

Like hell this is off topic. I want to know if the Dr. has a favorite color. The article specifically asks for questions.

I think the quickest way to show how artificial "she" is, is to show how hard it would be to answer a question like this.

OK, now let's see how artificial you are.

You obviously didn't read the blurb, where it's made perfectly clear that the doctor is human-- more specifically, a human philosopher studying the theoretical psychology of how humans will or may interact with androids or other artificial intelligences.

Since you did not read the article, and instead made a half-assed assumption based on your own faulty reading of the blurb, you must be a human slashdotter. To a painful extent, you are the real thing.

You also exemplify one of my favorite axioms regarding the topic: "Artificial intelligence is no match for genuine stupidity."

By the way, my favorite color is blue.

Re:Honest Question

[Anonymous Coward]

Does that question interest you?

Re:Honest Question

[FrYGuY101]

Ah, but you forget that different wavelengths of light can be used in different capacities. For instance, Blu-Ray disks are higher capacity than traditional red-laser DVD's, due to the shorter wavelength. If this were the case, they would indeed prefer blue for functional reasons...

Then again, I doubt a machine would perform significantly better than Eliza at the current stage in AI development, so the favorite color could just be if(rand(2)){return "blue";} else {return "red";}...

Re:Honest Question

[I_Love_Pocky!]

Ok, I just reread the posting, and I see that I misunderstood it. However, after rereading it, I still think it was very poorly written, it mentions no where that Dr. Pransky is a human. In fact, the way it is written seems to indicate that she is a robot (a robotic psychiatrist, not a human psychiatrist dealing with robotic issues).

It states that she is real, but it doesn't state that she is a real person, or a real robot. It says she is as qualified as any living human being to answer your questions, and I took this to mean that she herself wasn't a living human being (obviously I was wrong).

I was stupid not to follow the links and find out clearly what the article was about, but I think the only way anyone could know for sure what was indicated by the text of the posting would be to know who the heck Dr. Susan Calvin is. I have never read anything by Asimov, so I assumed Dr. Calvin must have been a robot who was also a psychiatrist.

Anyway, now that I'm sorry that I fouled everything up with my post, but I'm glad my mistake was dubed a troll (how kind). And thank you for so tactfully pointing out my mistake, as it proves without a doubt that you are clearly of a superior intellect.

Re:Honest Question

[I_Love_Pocky!]

I didn't actually forget that. I even mentioned it (violet light having higher energy than other colors).

It doesn't matter anyway, since I misunderstood the posting, and asked a question directed towards an AI, rather than some kind of wierd human psychiatrist that deals with human-robot interactions.

In regards to the Failure sensitive circuit:

[saskboy]

How would I coddle my robot in order to make it feel more loved? We all know that machines are most likely to break down when their Failure sensitive circuit is activated, so how do I show Robby that I care about it, but don't make it think that I need it to work?

She is as qualified as any living human being?

[Anonymous Coward]

no, thanks. I save my questions about robotic behavior for dead people. I don't have any questions about robotic behavior but that is besides the point.

Aren't you just another shameless tech self-public

[Sanity]

I spent a while looking through the "publications" section of your website to seek out the "hard academic underpinnings" that Roblimo mentioned, but all I could find there were a selection of puff-piece articles, vaguely gushing about a brave new robotic future (without actually saying anything that Asmov didn't cover years ago, but he did it with infinitely more elegance and forsight).

Which brings me to my question: Do you do any scientifically valuable research? I ask because you seem like just another shamelessly self-publicising cyber-pundit, much like the UK's Kevin Warwick (who, famously claimed to be the world's first cyborg after implanting a dog-tracking chip in his arm).

If not, how do you justify the damage people like you your supposed fields of research when your wild and glorious predictions fail to materialise? Aren't you just further widening the credibility gap between the promises and realities of artificial intelligence?

Morality

[Anonymous Coward]

Do robots consider themselves more or less "moral" than humans? Are they held to a higher or lower moral standard?

Re:Eliza

[Anonymous Coward]

You: Are we playing questions?

Eliza: Would you like it if they were not we playing questions?

One thing I'd like to know

[Elpacoloco]

Could a computer or robot be said to have a "mind" the way a human does?

What is the difference between "mind" and "software"?

Re:One thing I'd like to know

[fastdecade]

Could a computer or robot be said to have a "mind" the way a human does?

Define "mind" and I'll tell you if a computer has one.

What is the difference between "mind" and "software"?

Define software, and I'll use tell you how it differs from your definition of mind.

Not trolling, just demonstrating that this sort of deep philosophical questioning (which often happens in AI) usually just boils down to a trivial game of words.

Re:One thing I'd like to know

[hackstraw]

Could a computer or robot be said to have a "mind" the way a human does?

1st, you would have to define "mind". A typical definition goes like:

The human consciousness that originates in the brain and is manifested especially in thought, perception, emotion, will, memory, and imagination.

The key points here are "consciousness". In order for a computer or robot to have a "mind" it would have to be "self aware". Cognito ergo sum for the philosophy ppl out there. HAL from 2001 and 2010 became self aware and it caused problems. Skynet from the Terminator series became self aware and it caused problems. YMMV with any other self aware systems :) Btw, many phychologists do not belive in "minds", only brains, but that is an entirely different subject.

What is the difference between "mind" and "software"?

"Mind" see above. "Software" is a series of instructions to control a computer. I don't see any relationship between the two.

Re:One thing I'd like to know

[CaptMonkeyDLuffy]

Actually, as far as HAL goes, if I recall correctly, the reason for HAL going insane had nothing to do with becoming self aware, but was because lying and witholding information were against the 'nature' of hs programming, and he was instructed to lie/withold information(regarding the sighting of a monolith near Jupiter) from the crew...

Re:One thing I'd like to know

[hackstraw]

"Lying", "witholding information", and "going insane" are not computer problems. They are phychological problems generically labled as "cognitive dissonance". None of these would bother anybody (or anything) if they were not self aware. Something going against the 'nature' of his programming is a bug, it happens everyday, but I've never seen a machine act like HAL.

Re:One thing I'd like to know

[Verteiron]

This is why we have the Turing test. Simply put (as I understand it, I'm sure someone will correct me if I'm wrong) the idea behind the Turing test is that if you can't tell the AI from a biological intelligence, then by definition the AI must be on equal footing with the real thing. If, by talking to an AI, you cannot tell whether or not a human somewhere responding is actually to you, then the AI passes the test and must be treated with the dignity and respect due another human being.

In a related subject, many people try to argue that a computer or AI will never be able to actually "feel" emotions, only imitate them, but there's a fundamental problem with that way of thinking. You see, there isn't any way we can "prove" that you, or I, or anyone else actually possesses a "mind" or has emotions. I mean, sure, there may be the right chemicals in your bloodstream, the brainwave patterns might be right, but are you -really- happy/angry/upset/whatever? No way to prove it. Likewise, if an AI "acts" angry, and responds to all stimulus as if it were angry, then.. well, it's angry.

If the AI "acts" as if it possesses a mind... then it does, as much as we do, anyway.

Re:One thing I'd like to know

[Anonymous Coward]

and just how far is it between a BSoD and "i'm sorry dave, i can't do that"

Re:One thing I'd like to know

[An Onerous Coward]

The most common attempt at refuting the Turing test is called "The Chinese Box" argument, invented by Dr. John Searle. It basically goes as follows: Imagine that you're sitting in a room with one opening. The opening allows in tiny slips of paper covered in chinese symbols. You have a big book which you use to decode the symbols into other chinese symbols, which you dutifully write on another piece of paper, then deposit it in the opening.

To the outside world (which you are utterly unaware of), it's an amazing thing. They're holding a conversation in Chinese, by putting their words into the box, and getting coherent responses back out. But according to the argument, all that is really happening is mundane, mechanized symbol manipulation. The manipulator of the symbols (you) has no clue as to what they actually mean. The conclusion we are meant to draw is that elaborate symbol manipulation doesn't require actual comprehension.

I've never liked the argument. The whole thing totally glosses over one critical fact: in order for it to be passing the Turing Test, the rule book cannot be static. It has to change with every input, just as our brains are subtly rewritten as we talk to others.

For example, if I put in the box, "New York is called 'The Big Apple' because there is a mile-high apple sitting in the middle of downtown," a naive book would update its 'New York' symbol to include the apple, and its 'apple' symbol to include the possibility of mile-high apples. Meanwhile, a smarter book would update the symbol 'guy I'm talking to' to include the information "Compulsive liar." A static book, on the other hand, wouldn't change, so the input would always return the same output (most likely, "How does that make you feel?").

So this book, sitting in the box, is slowly altering its representation of the world as the inputs flow into it. This representation is one of the hallmarks of intelligence.

Another hallmark is awareness of context. If you're having a conversation with a five year old who just learned that Albany is the capital of New York, and you mentioned that you were going to New York on a trip next week, the most likely response would be "Albany is the capital of New York." I think kids do this instinctively, to check their understanding against others.

But it would be simple to mimic human conversation by the simple trick of parsing the input for symbols and then regurgitating facts known about that symbol. "You're Willow. You are my friend, and recently gay." If you're wise to it, this sort of Turing crutch doesn't mislead for long.

The same goes for Eliza, which tries to ask questions about the symbols it finds without actually comprehending them. "And how does Scumway Box with a side of feathers make you feel?"

In other words, in order to hold a truly believable conversation, the updating of the symbols in the book has to be done in a very accurate and sophisticated fashion which leads to an apparently accurate picture of the external world. This includes self-awareness on the part of the box, and an awareness of you as the interviewer. If this is happening, then it seems impossible to argue that the chinese box doesn't really "understand" the world, or that it's really "just" shuffling symbols around. Even though that is what is actually happening, the process is also creating a strong correlation to reality.

The problem I see with the Turing test comes from the false negative side. I imagine that a computer could be truly intelligent, and have a perfect picture of the world, but fail the Turing test because it either doesn't try to mimic humans, or does so in an unconvincing way. For example, I ask, "What are you?" and it responds, "I am a cluster of 4000 Intel processors in a basement at MIT." Or it gives me near-instantaneous answers to complex mathematical problems.

In reality, the Turing test is more widely applicable than most people think. All you need is att

Re:One thing I'd like to know

[Herschel Cohen]

May not seem to be related, but last week in the NYT Science section a report on Dr Crick, et. al. research on the sense of being. The major points were how few and how localized the firing of neurons may be the source of our consciousness. Moreover, at death we are at an end.

I extrapolated this to inferring that perhaps machine consciousness may be more easily achieved than has been thought.

Read it, see what you think.

About Human-Robot Relationships...

[MagiGraphX]

I've watched too much Chobits perhaps, but is it right for a human to fall in love with an artificially intelligent(and emotional) robot? Just a thought of what could happen...

Re:About Human-Robot Relationships...

[Neil Blender]

Is it right for a human to fall in love with an artificially intelligent(and emotional) robot?

Are we talking pleasurebot or aibo? It makes a difference, you know.

Re:About Human-Robot Relationships...

[mahdi13]

That was covered on an Outer Limits episode where this guy got a 'state-of-the-art' robot to help him after he lost use of his legs...then she (the robot) fell in love with him and got all psycho when he stared going out with a human woman

Re:About Human-Robot Relationships...

[chris_mahan]

Blade runner?

Re:About Human-Robot Relationships...

[compwiz]

heh, someone has been reading too much dieselsweeties...

Re:About Human-Robot Relationships...

[rcamera]

it happened in one of asimov's robot books, the naked sun which is about a human, gladia, and her murdered robot partner jander. good book, great series.

Re:About Human-Robot Relationships...

[Feanturi]

I've watched too much Chobits perhaps, but is it right for a human to fall in love with an artificially intelligent(and emotional) robot? Just a thought of what could happen...

If it's got the body of a Real Doll, then yes, it's perfectly right. Where is mine???

...self-publicist (subject line truncated)

[Sanity]

Doh!

gender diffs in attraction to the 'bot battles

[HealYourChurchWebSit]

Is there a big difference in gender between the audiences. If so, what is it about the battling 'bots that one sex find attractive over another? That is, are we looking at more hormonal/emotional causes, e.g. testosterone, or is there something intellectually more rewarding to one gender over another?

Wouldn't Battlebots be better…

[Anonymous Coward]

...if humans would enter the ring and bash each other over the head with remote controls?

2 words...

[BackwardHatClub]

Love Robot?

Future of robots?

[Merkuri22]

We've all seen the movies and read the books about machines in the future, and frankly most of these stories portray robots and AI as terrifying things that humanity will end up battling with for supremicy of the planet. Do you think there are any truths to these stories? Will robots compete with us in the future for jobs and/or living space? Do you ever see robots and humans living side by side as equals, or do you think they will always be subservient machines? Or, even, do you think robots will surpass us one day as the dominant force on the planet?

Re:Future of robots?

[Anonymous Coward]

I for one would like to welcome our new silicon overlords...

What form will A.I. take?

[mykepredko]

Hey Joanne,

A bit of a navel gazing question for you; what form do you think A.I. will take when somebody finally comes up with a program that is accepted as intelligent?

My own feeling is that the first A.I. program will simulate a simple life form (like a worm) instead of a highly complex and communicative form like humans. This goes against what Dr. Minsky believes A.I. should be, but I can't honestly believe that our first interaction with an intelligent mechanism would with something with similar capabilities to ourselves, but with something with the same mental capabilities and capacities as a bug.

The important aspects of Aritficial Intelligence will be making sense of its environments and learning from experience. To demonstrate that the Intelligence is learning is observing and testing the Intelligence's application of this knowledge.

What are your thoughts?

Thanx,

myke

Re:What form will A.I. take?

[Neil Blender]

something with the same mental capabilities and capacities as a bug.

Seems to me that most bugs act so predictably that their behavior is is hardwired and not really intelligence so to speak. I have seen documentaries where people have built artifical ants and observed the same behavior as real ants. Wouldn't you need interaction on a higher level to really determine if the machine is actually intelligent or just an illusion?

Re:What form will A.I. take?

[Tango42]

Worms are not usually considered intelligent beings...

Re:What form will A.I. take?

[Anonymous Coward]

Now, if I looked at the code for an artifical ant, could I set up a situation where it would not react intelligently? And if so, would a real ant react the same way?

Some people have gone after the Turing test by hard-wiring, basically they give the machine all the answers beforehand. The trick is that all you have to do is ask it something that's not on its list, and it doesn't have any way to respond intelligently. To qualify as intelligent, it has to be able to decide how to react to new situations (although with limited information and intelligence, the reaction may not be the right one.)

Personally, I agree with the grandparent. Crawl before you walk. Start with simple intelligence. If it simply scales to human level, great. More likely we'll slowly figure out how evolution created human intelligence as we repeat history ourselves. (I knew I should have attended lecture instead of working on my AI. Stupid general education requirement).

Proving that humans are intelligent, and not simply hardwired, is left as an exercise for the reader.

Re:What form will A.I. take?

[Zabu]

This is a good question to ask.

followup:
What sort of tests do you think are appropriate to determine the intelligence of software?

Re:What form will A.I. take?

[Laur]

Worms are not usually considered intelligent beings...

Yet they are often elected into high office.

Re:What form will A.I. take?

[indianajones428]

My own feeling is that the first A.I. program will simulate a simple life form (like a worm)

IIRC, we determine which species are intelligent by testing to see if they are self aware (and we test for self awareness by seeing if they can recognize themselves in a mirror). Last I looked, a worm didn't fulfill this criteria.

Although it's a step in the right direction, I don't think replicating a lifeform that isn't naturally intelligent could be called a true artifical intelligence.

Re:What form will A.I. take?

[Anonymous Coward]

The test for artificial intelligence is straight forward, and was devised by Alan Turing over 50 years ago. The criteria are simple: a human operator sits at two terminals, one of which is connected to another human, the other is linked to the machine being tested for AI. The operator enters into a discussion on any subject he chooses, through both of these terminals. If at the end of both exchanges the operator is unable to determine which of the terminals was linked to the the other human, then the machine being tested has true AI.

The mathematician Jon von Neumann theorised about artifical intelligence, and about the direction machines would 'choose' to follow once AI has beeen created. Amongst other things, he theorised that like all living things it will take steps to ensure its survival, and that it will replicate itself. Other scientists have added to his ideas, and some of the theories are catastrophistic in nature, claiming that the first decision an AI machine will make will be to ensure that it can't be switched back off. Science fiction maybe, but von Neumann was no crackpot, and was one of the foremost mathematical of computer science thinkers of his time. His ideas do have a serious following in sectors of the scientific community today. He predicted the creation of the Internet as a precursor to AI, and proposed that any AI machines would use the it to their advantage... perhaps to source materials for the process of replication.

If you consider it, a truly intelligent machine with the ability to access any machine anywhere in the world at anytime for any purpose will be as near to God as it is possible to get. A fascinating subject, but with questions that will only be resolved on the day that we finally boot up that machine!

Re:What form will A.I. take?

[scotch]

Good question. Here's another followup:

Imagine the following scenario:You're given a calfskin wallet for your birthday. What do you do?

Re:What form will A.I. take?

[Anonymous Coward]

I wouldn't accept it. Also, I would report the person who gave it to me to the police.

Re:What form will A.I. take?

[Anonymous Coward]

How the hell can this get modded up to +5? Has this guy followed ANY kind of AI ever? Has he ever played a FPS? Or any modern game for that matter. Check out Half-Life 2. I would say that the computer characters in games such as these ALREADY act smarter than bugs. Wake up dude (and mods). We are already far beyond bugs and worms in our AI design. When people talk about AI they want AHI, that is Artificial Humanlike Intelligence.

Re:What form will A.I. take?

[EverDense]

Worms are not usually considered intelligent beings...

Yet they are often elected into high office.

So THAT is what the "W" stands for.

Evil Robot Takeover?

[AvoidTheNoid]

How do robots ever get anything done if all they do is think up evil schemes to kill one John Connor?

2000

[Anonymous Coward]

Will I dream, Dave?

uh oh

[Valarauk]

As a freshman soon-to-be sophmore psychology major I'd like to know how she thinks her work will impact the current human counseling industry over the next 40 years...

Comment removed

[account_deleted]

Comment removed based on user account deletion

I would like to know.

[leastsquares]

Could my girl robot really learn to love me? ;)

Re:I would like to know.

[Saturninus]

No, not really.

I want to know

[maxpublic]

how soon we can expect a merging of realistic human-mimicking robots with RealDolls. And once that's done, will I be able to get my new humaniform RealDolls in the form of a blonde 15-year-old with a penchant for cheerleading outfits, or will the government ban this as some sort of cyber-pedophilia?

Max

Re:I want to know

[WormholeFiend]

i dont think anyone made a non-robot kiddie realdoll yet... so, you could take a cue from that situation.

Re:I want to know

[DarkHelmet]

or will the government ban this as some sort of cyber-pedophilia?

The question comes down ultimately not to what the robot is imitating, but ultimately, what is is.

There was a woman I used to talk to online a while back. She was 21. The thing was, in her pictures, she looked 12-13. Really young, skinny, asian girl.

There's a girl I talk to online right now. She's almost fifteen. But then, you'd think she's 18 if you didn't know better. There were a couple trolls a while back who posted pictures of her on slashdot, and they all drooled over her.

If the law is about what's perceived, then having sex with, or looking at naked pictures of the 21 year old would be considered pedophelia, even though she's of age.

On the other hand, if the law is about what's perceived, then having sex with, or looking at naked pictures of that 15 year old would NOT be considered illegal, because you'd be under the perception that she's older. In spite of the fact that she's faaar more womanly than the 21 year old.

Ultimately, a robot that would mock a 15 year old cheerleader is ultimately a 15 year old robot cheerleader. You should be able to do whatever you want to a robot...

... Until level of sentience becomes an issue. Robots with a true-to-human level of intellegence, crafted as 7 year olds for the purpose of a molestation toy is VERY disturbing. Humanity will have to cross that bridge when we reach it.

22nd century morality is a bitch. Makes me glad I live in the 21st... heh.

Re:I want to know

[ScottSpeaks!]

Most anti-porn laws try to have it both ways (and succeed). If the person in the movie looks like an adult and is presented as on, but is really a minor, it's kiddie porn and you're going to jail. If the person is an adult but is presented in a way that suggests he or she is a minor, it's kiddie porn and you're going to jail.

Likewise, as the tech to simulate under-age actors makes its way into the porn industry, I'm sure that "no actual minors were used in the production of this film" will not be a sufficent defense, either for the producer or the buyer. If the RealDoll folks were to produce a seemingly-under-age model - even one that just lies there - the bills criminalizing it would move through the state and federal legislatures so fast you could fly a kite in the breeze.

Re:I want to know

[DarkHelmet]

If the person is an adult but is presented in a way that suggests he or she is a minor, it's kiddie porn and you're going to jail.

Ultimately, define presentation... Would it be simply how he / she looks, or what other things she would do... I find that these are / should be ultimately irrelevant.

What if a woman who looks young chooses to do porn? Wouldn't her rights be violated in being of age, and not being able to do that work? It's a very interesting question, isn't it?

Re:I want to know

[ScottSpeaks!]

I doubt you'd get very far with a "right to be in porn" case in the courts. Not when weighed in opposition to the "public morals" question. (Not my opinion, just my analysis.)

Most porn (at least magazines and videos sold in the US; non-US web sites are another matter) includes a statement on the cover that "all actors at least 18 years of age" specifically to cover their asses on this point. I've also seen additional disclaimers on stuff that had the word "boy" or "girl" in the title, or which featured teenagers with (for example) pigtails and lollipops.

I agree that "what it looks like" should be irrelevant. The only question should be whether actual minors are being taking unfair advantage of. Mr. & Mrs. Christian & Prudence America seem to disagree.

The 3 Laws of Robotics?

[cy_a253]

First a reminder for everyone (Asimov's 3 Laws of Robotics):

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.

2. A robot must obey orders given it by human beings except where such orders would conflict with the First Law.

3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

Do you think it will be really possible to "hardwire" the 3 laws, (especially the first one) into robots? How?

And won't that require the robots to be capable of "abstract judgement", a quality only observed thus far in human beings? How could we implement that? Is it possible?

Re:The 3 Laws of Robotics?

[swamp_water]

What if Microsoft created robots with these rules. Would anyone trust that system not to crash at the worst moment?

Re:The 3 Laws of Robotics?

[Anonymous Coward]

Do you think it will be really possible to "hardwire" the 3 laws, (especially the first one) into robots? How?

Easy, a simple if statement will do (and I am assuming the first true AI will be written in Perl):

if ( $human ) {
$injure = 0;
}
else {
$injure = 1;
}

Re:The 3 Laws of Robotics?

[unknown_host]

You forgot an oft forgotten one .. The Zeroth Law

0. A robot may not harm humanity or, through inaction, allow humanity to come to harm.

Of course, now, the first law has a "except where suck orders would conflict with the Zeroth Law". This makes sense, since it allows it to make a decisive choice when stuck with having to sacrifice some humans to protect a larger number, where previously, it would not have the power to make any decision. Of course, all this is science fiction and still not fathomable with current technology and AI.

Re:The 3 Laws of Robotics?

[FreemanPatrickHenry]

Regarding Asimov's First Law ("1. A robot may not injure a human being or, through inaction, allow a human being to come to harm."), and assuming we "hardwired" robots with a "nature" in this sense--how would one (a robot) deal with moral paradoxes?

For instance, what if a robot is on a runaway train car that is about to kill five workers on one side of the track. However, if it pulls a switch, it can move the traincar onto another track. Doing so would save the five workers, but kill one. Does the robot make a qualititative judgement of human life? Does it decide "five lives are better than one," or does it try to decide which human life is "more worth saving?" (IE, if the one worker is a mother of four little children).

(This scenario was adapted from a recent Discover Magazine article on human morality.)

Re:The 3 Laws of Robotics?

[princewally]

According to one of the books, the 3 rules are not as simple as the text makes them out to be. There is a complex set of formulas for each of the rules.

Re:The 3 Laws of Robotics?

[Anonymous Coward]

In many of Asimov's stories, the robots simply went "insane." They were unable to deal with the paradox.

Re:The 3 Laws of Robotics?

[joebok]

There are a number of Asimov stories dealing with such things - the basic thing is that a robot will deal with it according to the sophistication of it's programming. Most would probably switch the car and just kill one. Those would most likely have their positronic brains fry on them for having taken an action that killed a human. Only a very few would be able to cope - R. Daneel for one. Giskard essentially made that same choice by saving humanity by killing humans (Robots of Dawn?). He invented the zeroth law of robotics, A robot may not through aciton or inaction allow harm to come to humanity, and then amended the first law with an exception for the zeroth. He didn't survive the process himself.

Re:The 3 Laws of Robotics?

[Anonymous Coward]

Bender would ask is there any way to get all 6.

Re:The 3 Laws of Robotics?

[javatips]

There is no concept of numbers of human being armed is the first law (the only concept ressembling that would be the law 0 that Daniel "invented"). There is some chance that the first part of the law will be evaluated first.

So like most human being (there have been many studies on the subject), the robot will probably choose to stay passive if this kind of situation arise.

Either way, like most human being, the robot will probably be messed up in some way and need the help of a robot psychologist to help him cope with the effects of his decision.

Re:The 3 Laws of Robotics?

[Anonymous Coward]

Luckilly for us, Bender is too lazy and shiftless to put the needed work in to "kill all humans".

Re:The 3 Laws of Robotics?

[Make]

what you describe is not artificial intelligence, but an expert system. many people mix these, although it should be obvious that "intelligence" cannot be expressed in a ruleset.

Re:The 3 Laws of Robotics?

[salvorHardin]

In one of Asimov's books (think it was Robots of Dawn), a robot had been constructed which still had all three laws, but their definition of what was or was not human had been altered. Therefore, they were able to harm/kill people who didn't have a certain accent, as without the local twang, people didn't count as being 'human'.

Re:The 3 Laws of Robotics?

[chris_mahan]

What would you do?
Don't ask a robot for a formula when even you can't provide one for yourself.

Re:The 3 Laws of Robotics?

[uberdave]

The "zeroth law" was not built into robots. It was an extrapolation made (I believe) by R. Daneel Olivaw and another robot.

Re:The 3 Laws of Robotics?

[old_unicorn]

Worse than that - if it could never allow any human to come to harm through it's own inactivity, then as soon as you switched it on, it would get a plane ticket out to the nearest drought zone, or earhquake, or war zone or.... and you would see millions of robot helpers digging and shifting and sitting on people's rifles. Even after all of the major calamities were dealt with, you couldn't have a beer too many - the robot woudn't let you, nor could you sit in front of the TV and eat pizza all evening - "One more slice only, remember your calories". Great, sanctimonious robots. All that technology and effort, and we have computerised mothers, (with an IQ of 5000). Maybe they'd 'let' us make some of our own mistakes, "as a learning experience", occasionally. Gee, thanks.

Re:The 3 Laws of Robotics?

[Lemmeoutada Collecti]

If I recall correctly, it was actually formulated by R. Giskard, R. Daneel's contemporary, and in as much as possible, friend. R. Giskard then used his rather unusual ability to modify R. Daneel's programming, imprinting the Zeroth Law onto the Positronic matrix.

In both cases, the hardwiring was not so much of the laws into the Positronic part of the brain. In Asimov's explanation, the laws were fundamental limitations in the architecture of the system.

Disclaimer: IANARY (I am not a roboticist yet)

Re:The 3 Laws of Robotics?

[SIWaters]

Have you not read Caliban? There is a "zeroeth" law, now: "A robot may not injure humanity, or, through inaction, allow humanity to come to harm."

Makes ya kinda misty eyed, don't it?

Re:The 3 Laws of Robotics?

[UserGoogol]

Yes it can.

Re:The 3 Laws of Robotics?

[Make]

prove. don't forget to attach your definition of "intelligence".

Re:The 3 Laws of Robotics?

[heydonms]

As I see it there are 3 answers, I think all have been valid at one point in Asimov's timeline. In one short story (i forget the title but its a bout a Calvin story about a robot that can read minds) Calvin traps the robot in a position where it can no act without breaking one the first law, it shuts down and "dies". In the short story "...That thou art mindful of him" USRBAMM start work on a robot for around the home and it has to start taking into account what it considers human (should it take orders from an idiot or a child), and a situation similar to what you describe is considered the result is that taking into account everything would waste too much time and the robot would fail to react so it goes by numbers 5 lives are worth more than 1. The third possibility as some one else mentioned is that it obeys the 0th law, as explained (created?) by R. Daneel Olivaw (sp?) where presumably the robot can take many of the variables into account and will act so as to "minimise harm to humanity"

Re:The 3 Laws of Robotics?

[Anonymous Coward]

Please actually go and read the books. Simply the ability to define and perceive "harm" and "human being" require a lot of subtlety, which Asimov's and his readers have been exploring for decades.

My question...

[hookedup]

Dr. Joanne Pransky, do you see Asimov's 3 laws of robotics playing a role in our relationship with robots in the future? Since most of our technological advances seem to come from developing warfare systems, will the 3 laws be left by the wayside, or will it become an integral part of robotics in the years to come.

Robotics vs. Programming

[swamp_water]

I'm a Robotic Specialist and I found I had plenty of Mechanical and Electronic skills when I left school, which was great if I wanted to repair assembly lines, but when it came to programming I had to go back to school to get more education. Do you feel Robotic people are lacking skills in computer programming and are behind computer people because of it? or more specifically, Do you think Computer Programmers are more qualified to build robotic systems then robotic people and thats why we have such limited robotic tech compared to the even today's video games?

Re:Eliza

[Alexis Brooke]

I think Eliza needs a bit more training:

You: Please do not eat my leg.
Eliza: Why do you say your leg?
You: Because I need it to walk on.
Eliza: Do any other reasons not come to mind?
You: It would hurt to have my leg eaten.
Eliza: Why do you say your leg eaten?
You: Because that's what we were talking about.
Eliza: Do any other reasons not come to mind?
You: You're gonna go ahead and eat it, aren't you?
Eliza: How long have you been gonna go ahead and eat it arent i?
You: Holy fuck, you're deranged.
Eliza: How long have you been deranged?
You: Since I was four.
Eliza: Why do you tell me you were four just now?
You: Aaaargh!!
Eliza: Please go on.
You: No.
Eliza: You are being a bit negative.

WyldStalyns!

[AvoidTheNoid]

Who would win in this fight?

Good Bill and Ted Robots versus the Evil...but far better built Evil Bill and Ted Robots

The obvious question:

[Bingo Foo]

10 INPUT ANSWER$

Uncanny Valley

[Anonymous Coward]

With the development of increasingly humanoid robots, such as QRio, Asimo and David Hanson's KBot, what are your thoughts on what will happen when we reach the 'Uncanny Valley'? Will it have an effect on robot-human interaction? Or is the 'Uncanny Valley' theory completely false?

The Subject of Choice

[beeglebug]

Can you forsee a point at which intelligent machines/robots will refuse to allow humans to program them any more? If so, how will this affect society?

I don't necesarily mean in a malicious way either, just that at some point artificial intelligence might advance to the point where it would percieve human intervention as potentialy damaging, and respond accordingly.

Re:The Subject of Choice

[ziggamon]

It will go a little something like this:
First, they will learn to use the sun for power.
then, people will have to block the sun to protect themselves from the robots.
then, the robots enslave the humans, by putting them into red egg-shaped containers, using their body for energy. The humans themselves will be connected to a dream-like digital system, which will give them an impression of living a life like we do now...

In the end, there will be some dude that learns to fly like superman by "hacking" this system...

Human Features of Robots / Bonding with robots

[jhouserizer]

Over the years, there has been a fair amount of debate about whether robots should take on human forms, especially with regards to having detailed life-like faces. Some robot designers, wary of this debate, have settled on giving their creations near human-like faces.

My question is in relation to this topic. Do you think that people (and "sentient robots" that may exist some day) will be be overall better served if robots are readily distinguishable from humans? How strongly will this affect our "bonding" with robots and their bonding with us? Dogs for instance look quite different from humans, but many a family-pet seems to believe itself to be a real part of the family, and sometimes even seem to think themselves to be human. How will this affect the way we deal with "death" of a robot?

Three Laws of Robotics

[Barkmullz]

Do you think future robots will adhere to the Three Laws of Robotics, as defined by Isaac Asimov?

Cyborg vs. Robot

[arjay-tea]

Why are techie types so heavily drawn to fully automomous robots, virtually ignoring the vast potential inherent in the cybernetic enhancement of already-formidable human faculties?

Artificial intelligence without embodiment?

[macshune]

Dr. Joanne Pransky,

As an undergraduate philosophy student interested in the theoretical implications of A.I., could you tell me what your thoughts are on the validity of the assumption that artificial intelligence is possible separate from the notion of embodiment? I think the lack of consideration given embodiment is one reason why artificial intelligence researchers have come up empty-handed so far in their quest to synthesize a conscious, self-reflective entity.

To ask the question more succinctly, do you think a mind needs a body and possibly and environment to interact with in order to be conscious, or can a mind exist and know itself independent of an external context?

Re:Artificial intelligence without embodiment?

[jbrader]

Damn I wish I'd thought of that one. Somebody mod this guy up. I'd do it myself but I already posted.

Re:Artificial intelligence without embodiment?

[shams42]

Wow, I wish I had mod points so that I could mod you up. I've been thinking this very same thing for a long time now. It seems to me that the "brain in a vat" conception of AI is likely to meet with continual failure.

Re:Artificial intelligence without embodiment?

[Anonymous Coward]

I don't think a physical body is required to provide a context. The Internet comes to mind as a particularly rich environment.

Of course, the resulting "conscience" wouldn't be anything like a human being, but what's the point of copying something like that. :)

Re:Artificial intelligence without embodiment?

[Anonymous Coward]

Not to flame, but this questions shows a fundamental mis-understanding of the focus of AI research. AI research is essentially focused on the design of a systems that can react in "intellegent" and "responsive" (soft words) ways to given stimulus. In the ABSENCE of an environment, sense or context, most AI systems are not designed to function. The whole point of the research can be boiled down to: How do we sense our environment accurately and, that done, how do we respond to it meaningfully.

No one is trying to develope "brains in a vat" scenarios. The "body" the AI inhabits is a fundamental consideration when developing sensing/reacting mechanisms. In the absence of these considerations, the AI system has no input and can produce no output.

Re:Artificial intelligence without embodiment?

[Anonymous Coward]

Was there any genuine reason whatsover in your mentioning your degree?

No, I thought not.

Robot Emotion - What's the Point?

[fastdecade]

Humans certainly have a range of emotions - is this an evolutionary advantage to be injected into robots or an inefficient side effect to be disregarded?

3 words...

[RCO]

Blow Up Doll!

Honest question, can i get an answer?

[c5r]

What are the main differences in the way (ways?) a robot sees within its own physical construction and operates that physical system optimally and the way (ways?) a human sees within its own physical construction and operates that physical system optimally?

From the dept.

[Anonymous Coward]

Which dept? I don't get it.

Old Joke

[niko9]

How many psychiatrists does it change to change a light bulb?

One, but only if the lightbulb really wants to change. ;)

--

Not to be a Dick, but....

[LittleGuy]

Do Androids Dream of Electric Sheep?

SW:KOTOR

[WormholeFiend]

In Star Wars Knights of the Old Republic, there is this woman who, um, has an unusual relationship with her droid, after the death of her husband, which leads the droid to want to commit suicide...

Is this likely to happen in the future? I mean, the unusual relationship, not the robotic suicide.

How would you treat such a dysfunction?

Re:SW:KOTOR

[gevmage]

Just for the record, in one of Asimov's robot books (I forget which one) this happens as well. A woman has an intimate relationship with a humaniform robot.

She wanted it to (thus satisfying Law 2) and it made her feel good (indirectly satisfying Law 3).

Dr. Sbaitso

[funk_phenomenon]

Anyone remember Dr. Sbaitso? I used to play play around with that for hours. Didn't help me at all.

Re:Dr. Sbaitso

[SlipJig]

OMG - I had no idea it was out there for download! There's nothing like a bit of nostalgia (sniff.. sniff ;)

What about the cats?

[Anonymous Coward]

How long before these robots are able to catch and pull the heads off of cats? If we don't write some feline protection into the 3 laws, the stench of headless cat bodies will be enormous.

Re:Aren't you just another shameless tech self-pub

[Otter]

I'd never heard of her before and only know what's on her site, but -- she seems to actually be a marketing person with a relatively long track record in the robotics industry.

But, yeah, I have pretty much the same reaction you do to that "robot psychiatrist" shtick. (Roblimo definitely seems to prefer arranging interviews with various freak shows than with dull people with real accomplishments.)

Absolutely!

[Hobbex]

Anyone who gives a ton of interviews and appears all over the press talking about the revolutionary promise of some technology that never quite delivers ought to be ashamed of themselves!

Right? :-P

Re:Absolutely!

[Sanity]

Anyone who gives a ton of interviews and appears all over the press talking about the revolutionary promise of some technology that never quite delivers ought to be ashamed of themselves!

Don't make me revoke your channel operator status you turncoat!

Re:Absolutely!

[elmegil]

Methinks you're too subtle for some of the zealots around here to quite gather.

AWESOM-O 4000

[stratjakt]

When will the pleasure model be made available to consumers?

Would you ever fuck a robot?

Can you build..

[RailGunner]

the Awesom-O 5000?

/South Park

awesom-o

[Anonymous Coward]

Awesom-o you're my bestest friend

A judge on BattleBots?

[digitalamish]

How can you call yourself a Dr. and just sit idly by while humans force their creations to battle to the death for sport? Where do you draw the line between 'just being a robot' and being a 'slave'?

---
"Have the lessons of Terminator been lost on all of us?" - overheard during trailer of I, Robot

Re:A judge on BattleBots?

[Anonymous Coward]

Quite an experience to live in fear, isn't it?

That's what it is to be a slave.

Sex?

[mikeophile]

Will extramarital sex with robots of various levels of sentience be considered "cheating"?

Re:Sex?

[cozziewozzie]

And if she didn't know that this interview was for Slashdot, this question would make it more than evident...

Re:Sex?

[K8Fan]

Will extramarital sex with robots of various levels of sentience be considered "cheating"?

Given that a huge number of women consider their husband or boyfriend watching porn and masturbating by himself "cheating", I think we can safely assume the answer is "yes". Sue Johansen's show "Talk Sex With Sue" deals with that "issue" nearly every week - some woman calls in freaked by finding her boyfriend/husband's secret porn stash. Humanform sexual robots would definitely be considered cheating. I'd venture to guess that they'd be even more freaked out, because a well designed robot could have a level of physical perfection and skill that no human woman could possibly match. The true Turing test would be, not mearly whether a human can communicate with an AI, but whether a human could fall in love with one.

Re:Sex?

[xxdinkxx]

The true Turing test would be, not mearly whether a human can communicate with an AI, but whether a human could fall in love with one.

I can see the politicans now "we must stop this abomination before it becomes reality"..I can also see how there could be a "marriage is between humans(or if the right wins between male and female)" debate.

Re:Sex?

[Anonymous Coward]

It's not real "sex" if you fuck the robot in the ass...

Re:Eliza

[br0ck]

In double-blind studies it has be found that people could not tell the difference between the automated psychotherapists Eliza and a human therapist. The same study showed equal mental health healing rates using either therapist.

Dr. Calvin, how soon until you lose your job to Eliza?

human-robot interaction

[Fratz]

It occurs to me that there may be technology to make robots appear to be human before there is technology to make them act human. Do you feel there's a need to pressure the industry to make sure their robots only appear as human as they behave, so that people do not have incorrect expectations about what the machines can do?

Um .. Grad Students?

[hashinclude]

How about Robot Grad Students?

Gyromite

[Craptastic Weasel]

So.. on level twelve, where the good sleeping doctor is walking between the the first doorway and the doctor squishing device (the one the good folks at Nintendo programmed into 2 seconds later in the game), I'm stuck.

Do you think it is even remotely possible to get that spinning gyro from the thing that keeps it spinning to the red button on one side, and then to the other side before the doctor meets his ill fate?

yeah... sigh.. me niether.... half to go back to cheating and hitting the button with my finger.

Re:Eliza

[Anonymous Coward]

You: shut your hole bitch Eliza: What about your own hole bitch? I've been pwned! :(

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

"...dull people with real accomplishments..."

Um, that would be the audience...

Roborights?

[jrpascucci]

Do you believe there will come a time that we will have a 'robot rights' movement? Will it be more credible than most of the 'animal rights' movement, or just a good-hearted (but weak-minded) anthropomorphization of our silicon companion machines?

Someone (Dennis Miller?) once said, animals can have rights as soon as they accept responsibilities. Robots obviously can be given responsibilities (your job is to fit tab A into slot B), but ethically, should they get rights? As soon as someone programs a robot to pass the turing test, and then immediately ask for his rights? Or is it something deeper?

Beyond some kind of second-class entity status, will robots become citizens? Do robots have a god-given right (recall, our rights are considered by the Declaration of Independence to be given us either by 'Nature's God' or by their 'Creator') to freedom of expression, association, religion? The right to bear arms? Do robots have a 'right to work'? "One Robot, One Vote"? Will Robots have to file tax returns? Will there be Robot Courts? Robot Lawyers? Robot Jail? Robot Schools? Robotic Members elected to the Legislature? Some day, will we have a Robot President? Is a Robot built in Japan eligible to be president? What if the robot was shipped from Japan as parts with software, and put together here, does that count?

If you start building a robot, and decide to stop, will that be considered to be a robaboration? Or the work of their 'creator'? And if, after building, you switch it on and then decide you don't like it that much, and power it off again and harvest the parts, is that robomurder and disrobomemberment?

-JRP

Re:Roborights?

[forkazoo]

As a side question, in relation to the parent... Assuming robots eventually are sentient, members of society, have a reasonable assortment of rights. What about recreation? What happens to society when robots realise that fleshies are building their kind solely as a labor force? Is there any chance that robots would try to make robot production a solely internal matter? What happens when the first research robot decides to clone a human or human like child, and raise it? Obviously, the question of robot production in this sense is a very distant one, so much so as to be a pretty moot point in our lifetime, but I still find it an interesting muse.

Re:Roborights?

[Hatta]

More importantly, will robosexuals be discriminated against? Will robanukah be an official holiday? Is there really a robot hell?

Re:Roborights?

[euxneks]

Just to add to this:
Obviously the Robots will not have any rights unless we have given them a true Artificial Intelligence. Once we have gained this monumental feat (say, an intelligence on the level of our own in an autonomous body) What sorts of rights does this entity have? Should it have rights?

Re:Roborights?

[Obfuscant]

Robots obviously can be given responsibilities (your job is to fit tab A into slot B),...

This is not a "responsibility" in the sense of "accepting responsibility". This is just a task.

What's the difference? If a human chooses to "fit tab A into slot B" and by doing so causes harm to someone, then "accepting responsibility" means that the human accepts whatever the consequences are for that action. For example, on the assembly line, someone gets between the tab and the slot when the human is doing the task, and he does it anyway. The interloper winds up skewered by tab A. Who is responsible? The human who didn't stop doing his task. (If the imagery does not work, imagine the huge welding robots that have replaced human welders, and make the task "weld part A onto part B.)

Would you punish a robot for doing the same thing, or would you punish the programmer/owner? How would you punish a robot? If it aint' smart enough to "accept responsibility", then it ain't smart enough to punish -- which is the basis for the diminished capacity defense in human criminal trials.

As for animals accepting responsibility, try this. The cow does not feel any responsibility to give milk to keep humans fed. The dog, as smart as he is, doesn't accept responsibility for keeping his owner from crossing a busy intersection, he is trained to do this and didn't have a say in the matter. Would you put the dog to death if his actions led the blind master into the path of a truck, or would you simply retire him from guide-dog service?

Re:Roborights?

[Wes Janson]

Didn't you ever WATCH The Matrix?!? /galaxyquest

Re:Roborights?

[c4ffeine]

This will probably be a problem if we augment, repair, and eventually replace our bodies with much more sturdy, capable mechanical bodies. It sounds far-fetched, but look: we have artificial joint replacements and are working on cybernetic arms, what's so hard about a whole body?

Anyways, this will lead to an interesting question: What is the difference between a human-like intelligence in an artificial body (a sentient robot) and a human intelligence in an artificial body/

Re:Roborights?

[Feanturi]

My cat is way more intelligent and aware than anything we are even close to building. My cat has practically no rights in our society (laws against animal cruelty are all she has). I don't think we're going to be able to answer questions of Rights and whatnot until artificial minds are much closer to our level. It's going to be a play-it-by-ear type of scenario.

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

I would have to second this, there are alot of "glam and glitz" intellectuals who pitch to popular audiences (howard rheingold is another example) rather than teaching courses and furthering research with their peers;

How much money do you make at your speaking events?
What are your main sources of income?

I will be very disappointed if the editors decide not to send the parent question in. I think although very forward, these are questions that need to be asked.

The best question...

[Cytlid]

I wrote an AI program back on my C64 as a teen, that tore apart sentences (and questions) and tried to derive the meaning of them from a database. The idea was I would add more info to the database, and sooner or later it would learn by itself and add to the database. The idea never got off the ground, but I did try with a quick small database, and asked it a quick question (which would be my submission):

Who are you?

(To which it replied "I am I" ... technically correct but totally useless.) Always wondered how a real robot would answer that...

Re:The best question...

[drox]

"Who are you?"

Always wondered how a real robot would answer that...

A real robot would be too busy whupping the silicon out of all the other robots in the arena to answer philosophical questions like that.

Re:The best question...

[hackstraw]

Who are you?

(To which it replied "I am I" ... technically correct but totally useless.)

Isn't "I am I" the best answer to "Who are you?" Most people, I guess, would answer with their name, "I am John" or whatnot, which is technically incorrect (how many John's are there?), and totally useless.

Questions like this are beyond 99% of the human population, and off limits to things that have no concept of "self".

Royal rumble?

[weeboo0104]

Who would win in the Battlebots arena,
R2D2 or TWIKI from Buck Rogers?

robot origins

[Anonymous Coward]

where do robots come from? if a daddy robot and a mommy robot fell into a mud puddle, would three robots come out?

My Question, of course.

[TailGunner]

Doctor, do you think it's fair if a robot watches a person after the person does something the robot can't copy? Just wondering..

walking, talking robots

[basil montreal]

It looks to me like a combination of the work being done by Sony and other Japanese manufacturers will give us walking machines that have the same type and degree of mobility as a human eventually. Also, the work being done in a university in Europe (Sorry, I can't seem to find the link anywhere yet, will go and reply to this when I have found it) seems to indicate that we may eventually have a computer program capable of holding a perfectly believable conversation with a human.

Do you think that the combination of these technologies will lead to walking, talking robots?

the canonical : IIIaFi%x%DIMMa%y%!

[Anonymous Coward]

If I install a Fleshlight in my Aibo, does that make me a PUPPYFUCKER?!!!

And where can I get the device drivers for *BSD?!!!!

Re:the canonical : IIIaFi%x%DIMMa%y%!

[Anonymous Coward]

To answer thy question: No, but I'd like to watch!!!111

Human LIke Robots

[QuantumFTL]

Assuming that some day, we eventually develop human-like android robots, do you feel that individuals who are unkind/abusive to these robots (regardless of whether or not they actually have feelings) are going to start treating other humans this way? If so, does that mean that there should be rules against abuse/cruelty of human-like robots, as a preventative measure against it happening to a real person?

The existence of "disposable people" would have to cheapen human life in the eyes of some. Are there any other problems with this? Is there anything we can do to prevent this?

Cheers,
Justin

Social implications of mass produced robots?

[Zabu]

This is a multipart question.
If robots are mass produced to carry out simple but time-consuming tasks in the future and are cheap enough to eliminate the need for a large percentage of the human workforce, do you think that there will be widespread anti-robot sentiment?
When human's jobs are replaced by a cheaper alternative, they feel a great injustice.
Do you think that robotic 'slaves' is really what an ever expanding population needs? Or will the creation of robots take a different direction to carry out tasks that humans cannot?

Re:Social implications of mass produced robots?

[Phoe6]

When the question of Mass Production has come, it needs to be observed that all the technologies of the past which have come to the mass production level had it's intial surge to that level from sex related things. Take for eg. Printing Press, Telephone and Internet (Prono). Will the same logic apply to the mass production of Robots as well? What will be the social implications if the same industry trend of Mass Production follows with a technology as Robots? How will be the 3 laws of RObots be fitting in this scenario?

Re:Social implications of mass produced robots?

[Anonymous Coward]

Looked at a Detroit assembly line lately? I think this is already happening.

Re:Social implications of mass produced robots?

[Wyzard]

If you like anime, you might want to watch "Armitage the Third". It's not deeply philosophical or anything, but it explores this situation from the point of view of a humanoid "third-type" robot, who, unlike the "second-type" robots which perform menial tasks, has emotions and a personality, and suffers from the anti-robot sentiment. (She also happens to be derived from a combat design, so she can fight for what she believes in, but that's sort of a different side to the story.)

The Return

[Saturninus]

Dr. Sbaisto?

where's the positronic brain?

[futuretaikonaut]

In Asimov's robot novels, the assumption was that modern science had invented the positronic brain, which was thought to be capable of actual sentient thought, though most of the robots in the books did so on a very basic and childlike level. It was this that actually gave Dr. Calvin a job... seeing as how the brains had the capacity for original thought, even though it was mostly predictable. As it stands today, and into the foreseeable future, we have invented no such thing capable of acting with original thought. Our hardware has, instead, given the appearance of thought, as it is capacble of so many calculations per second that it appears to come up with things on its own.

So, my question is, what use is a robot psychologist if every action that a robot can take is already predetermined by its programming? What new field is there to be discovered that is not already known? In the human mind, we are constantly learning new things about the brain, a mechanism we only barely understand, but what is there to derive from a machine we ourselves create?

Perhaps a better study would be the eventual effects on human society. A million questions remained unanswered regarding that.

Re:where's the positronic brain?

[euxneks]

So, my question is, what use is a robot psychologist if every action that a robot can take is already predetermined by its programming?

It sounds like you believe that we as humans have entirely undeterministic actions. That's neither proven nor unproven. Can you guarantee that a number you choose "at random" _is_ actually a random number? I doubt it.

Re:where's the positronic brain?

[An Onerous Coward]

Bah, nobody understands determinism.

I doubt there is any such thing as a "positronic brain." Not just in the literal sense, but also in the wider "magic box that is somehow endowed with the ability to perform 'original thought.'" Some forms of hardware (neural nets, grid computers, etc) may be more suited to intelligence than the processor->cache->RAM we're used to. But any Turing-complete architecture can emulate any other Turing-complete architecture. So no, it's not about finding the right style of hardware, nor is the current sort of hardware automatically disqualified from possessing intelligence.

To answer your question, what good is a human psychologist if a human is governed strictly by the contents of its programming (that is, its brain)? Personally, I think that our behaviors are in fact deterministic. The inputs that pour into our brains through our senses enter in a deterministic manner. They interact with its current contents according to the laws of physics and chemistry, strengthening and weakening neural connections. As the brain changes, the "us" that is represented within also changes. We can't choose to override the signals pouring through our brain, because those signals themselves are how we make decisions.

Some would argue that there are options for non-determinism: quantum indeterminacy, for example, could bubble upwards and create macroscopic effects. But that's not the same as "free will." You would have to postulate some external entity (a soul) that caused the indeterminacy to act in such a way that the result represented "your" wishes. Otherwise, quantum indeterminacy is just a sort of random number generator, creating variety but not intentionality.

So everything is going to happen the way it's going to happen, and therefore nothing matters, and you may as well jump in front of a bus. Right? Wrong. The most obvious fact is that your jumping in front of a bus makes the world a different place. If instead, you went on to be a Greenpeace activist, or a businessman, or a CIA assassin, you've made a difference.

But beyond that, consider science. Scientists keep performing experiments, knowing full well that their results are going to be absolutely and unequivocally determined by the physical and chemical laws governing the setup. Knowing that our own actions are a manifestation of physical laws tells us nothing about what those actions are going to be.

As an avowed determinist, I don't know what's going to happen today, or feel any sort of fatalism towards it. My decisions affect the world in unimaginably complex ways and the world affects me in ways I don't understand. My life will be different depending on whether my next act is to do my homework, browse Slashdot a while longer, or have lunch. I still haven't decided. But I do accept that--just as my understanding of the world is inherently incomplete--my understanding of myself is incomplete because the "I" I feel is only a representation of the real me.

I think that pointing out that robot action is predetermined by its software isn't a terribly helpful point to make. It's kind of an observation of the "water is wet" variety.

Re:What form will A.I. take? A computer?

[c5r]

Seems to me that I interact with 'intelligent' computers regularly. I receive mostly images and sounds from the computer, and reply with mostly digital feedback via my input devices. The whole concept of intelligence presently seems so wrapped around itself for a definition that I'd imagine acceptance as 'intelligent' would vary widely from culture to culture, person to person, and day to day.

Battlebots and the American Psyche

[Ransak]

I, like many people, really enjoyed Battlebots. So much in fact, I built one just like much of America thought about doing. What drives the fascination with Americans and the desire to build/tinker things that are capable of destroying each other? Other robotic competitions like FIRST are about completing tasks or doing something constructive (which I suspect is driven by a different motivation) while the more sensational tournaments were about robots killing robots. Is this just the desire to compete in 'left brain' individuals, or something else? And what makes competitons like Battlebots and Robotwars appeal to the American public?

Re:Aren't you just another shameless tech self-pub

[dj245]

Do you do any scientifically valuable research? If not, how do you justify the damage people like you your supposed fields of research when your wild and glorious predictions fail to materialise? Aren't you just further widening the credibility gap between the promises and realities of artificial intelligence?

The number of questions per post shall be Three. No wait, one. One Shall be the number of questions per post. The number of the questions in any one post shall be one. Two shall the number of questions not be, and under no circumstances shall the number of questions be three.

How many questions per post? Three. No wait, One. I mean, aaarrrgh! (cast into a deep ravine)

emergence

[shams42]

How do you think our emerging understanding of emergence and self-organizing systems will influence AI research and development?

I ask this because I have long thought that the mind or consciousness is an emergent property of the biology of our nervous systems.

6 words...

[Patrik_AKA_RedX]

Blow Up Doll Filled With Helium!
(So it can play "hard to get")

Re:6 words...

[DoomHaven]

Thanks for the laugh!!! :)

Choose your words carefully!

[vaxer]

Dr. Pransky is clearly not "an undergraduate philosophy student", so she cannot answer your question as one.

Human-level AIs will probably learn in conversations with people. Remember the implied lesson in SpaceCamp: be careful what you say!

Re:Choose your words carefully!

[macshune]

my bad! where were you when i was writing my post?:) apology noted in child post to grandparent. thanks!

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

I will be very disappointed if the editors decide not to send the parent question in. I think although very forward, these are questions that need to be asked.

That is exactly what I was thinking. For me, this is now a test. One of the reasons television stinks is that this type of question often goes unasked. Is Slashdot any better than television in this respect? I guess I will have to wait and see.

more than just robots

[ChipMonk]

What's the poorest example (preferably current) you've seen of human-machine interaction? What can we learn from it? I'm thinking specifically of GUI's, but I know there are probably zillions of bad HMI's out there.

I had an old Dodge Charger that was very poorly designed inside. Some gauges were blocked from view by the steering wheel, the heater controls were nearly out of reach, the radio was even farther away... It made me appreciate my Honda Civic CRX that much more when I got it.

C3PO

[Furan]

What problems would you diagnose the fictional Star Wars character C3PO with?

Re:C3PO

[Professr3]

What problems would you NOT diagnose the fictional Star Wars character C3PO with?

Re:C3PO

[ScottSpeaks!]

Threepio's problems stem mostly from his envy and resentment of Luke and Leia. Despite being his "father's" eldest surviving child, and even with Anakin's increasing mechanization (giving them more in common), Vader never paid any attention to him. It was all "You are part of the Rebel Alliance, Princess," and "Luke, you can destroy the Emperor." In Episode VII, he'll probably murder them both in their sleep.

Re:C3PO

[Anonymous Coward]

two words. Latent homosexuality. ;-)

Re:Aren't you just another shameless tech self-pub

[nharmon]

Going even further, I am curious what "Dr" Pransky's degree is in. She calls herself the world's first robotic "psychiatrist". Well, real world psychiatrists go to medical school. So aside from being experts on how the mind works, they also know quite a bit about psysiology, and biochemistry. Funny, Roblimo says we need to leave out the hard-tech questions. Why? If psychiatrists are doctors the same as any other, than a robot psychiatrist should be an engineer the same as any other.

Maybe she didn't go to Medical school. Real world Psychologists have graduate degrees in the field of Psychology. Since she calls herself a Dr.m, I'm assuming she finished a PhD (if she didn't attend medical school). What was her dissertation about?

What a scam it is when slashdot helps some chick stroke her ego and doesn't have the credentials to back it all up. Of course, we have unfortunately come to expect this from /.

Some approachable reading on the topic...

[aussersterne]

is Descartes Error, by Antonio Damasio. It's quite readable and presents a case for the conscious self (including emotions) being, at least in part, a manifestation of, or mechanism for representing, the ongoing state of the body.

SBAITSO.

[mikeleemm]

Only one comment I can make is Creative Labs, Dr. Sbaitso.

Re:SBAITSO.

[BMonger]

Hello, my name is Doctor Sbaitso
I am here to help you
Say whatever is on your mind freely
Our conversation will be kept in strict confidence

(of course this *was* in all caps but the lameness filter kicked in)

Quick answers:

[Anonymous Coward]

No, no, no, no, and no. Build robots so they can't suffer or feel pain. Then they can be turned off or killed with less guilt than accidentally stepping on a millipede, like I did this morning. :-(

Your favorite fictional robotic character

[Strange Ranger]

What is your favorite robot/cyborg character in written or film fiction? Why?


For instance, I'm happy to admit mine is Data from Star Trek: Next Generation. Most especially the earlier seasons. Reason: I'm not much of a "trekkie" but that character made me consider so many different possible aspects of AI and of being not-human. From trying to understand other humans' emotions to his contrast with 'The Borg' down to what it might be like to have an "internal chronometer". For totally different reasons I loved Douglas Adams 'Marvin the Depressed Robot' in HHGTTG.

Re:Your favorite fictional robotic character

[PatrickThomson]

Ahem. Paranoid Android.

Re:Your favorite fictional robotic character

[Anonymous Coward]

The problem with Data is mainly that Brent Spiner got fat. How does a android get fat? How do you explain that away without sounding absurd?

Whoops!

[macshune]

I meant that I'm an undergraduate philosophy student. Apologies to Dr. Pransky!

Useless question.

[aussersterne]

These are judgments even humans are unable to make cleanly or clearly. Entire panels of professional medical ethicists are routinely unable to agree on whether this or that process or product harms or hurts humans, which humans, and whether that harm or hurt protects the existence of the species in the long run or sabotages it.

Medical technology, genetically modified foods, physician-assisted suicide, abortion, the spread of electronics-based technology, nuclear power, invasion of Iraq...

This is basically Ethical Paradoxes 101; before we can program this sort of thing into machines, we'd have to be able to reason it all out ourselves!

Re:Useless question.

[old_unicorn]

This question - 1 dead vs 5 dead is relatively easy. The REALLY hard questions are where even the quantity of suffering is uneven, i.e. "everyone pays a little more tax, so that a few people get more help", "a few people pay a lot more tax, so that a lot of people pay less", "lots of people stop playing with guns, so that a few less people get shot" etc. These are moral questions which become political ones. Robots in politics, now there's a thought. "Watch my loudspeaker..."

Re:Useless question.

[Wittgenst3in]

>This question - 1 dead vs 5 dead is relatively easy You certainly seem to think it is. Consider this scenario: 5 asshats go to rob a bank, whilst escaping in the car they smack into another car at an intersection. Both cars turn over, the tanks puncture and spill over the cement. You (a paramedic) arrive on the scene with the ONLY jaws of life. Both the one innocent man and the 5 robbers are having their flesh burned off and screaming. Which one do you rescue first? I'm ashamed of this post because I think there should be a version of godwins law for 'trolley problems'. http://en2.wikipedia.org/wiki/Trolley_problem

When does a computer have feelings/emotions?

[langeland]

We all know about the Turing test, which suposedly (in numorous editions) are meant to tell wheather a computer program is intelligent or not. What about feelings or at least emotions? Do You have any criterions that distinguishes non-emotional/non-feeling computer software from emotional/feeling computer software?

Re:When does a computer have feelings/emotions?

[langeland]

Self citation I know! The question in it self may be lame if you are unaware that:

The cognitive sciences have unveiled a lot about the emotional system, which, in humans, is inherent from the evolutionary lower specieses such as flies, worms, snails, rats, cats, dogs, primates (including humans). Fear conditioning has been reported on all these (and many more) specieses. Yes, we seem to share the same neurological curcuits as those animals in our emotional system.

The functionality of the mechanisms behind emotions and behavior are now that well understood, that scientists are able to constuct computational models of the relevant areas of the brain. The most relevant areas of the brain implementing the emotional system are amygdala and orbitofrontal cortex (or ventromedial prefrontal lobe) depending on a general view on animals or humans respectively. It won't be long until computational model based on these findings make their way into applied AI. So the question is wheather agents that are implememted with these models will have emotions/feelings or not?

Pushing or Shoving?

[hoggoth]

In your experience with robots, which is the real danger, pushing or shoving?

Searching for God?

[b-baggins]

Do you think the push for AI in robotics is an attempt by people to find God? A being with all the human virtues and none of the human foibles that will come and bring utopia to our world?

M-x doctor

[joib]

I'm sure doctor-mode beats this thing.

SmarterChild

[Anonymous Coward]

I said the same things to SmarterChild, a moderately useful AIM buddy:

me: Please do not eat my leg.
SmarterChild: Don't eat your leg? Why not?
me: Because I need it to walk on.
SmarterChild: What?
me: It would hurt to have my leg eaten.
SmarterChild: There's no way it would hurt to have your leg eaten.
me: You're gonna go ahead and eat it, aren't you?
SmarterChild: I're gonna go ahead and eat it. Well that's interesting.

At least it gets the weather right (and faster than a web browser).

Re:SmarterChild

[Fjord]

Same deal, but with ZolaOnAOL.

Fjord: Please do not eat my leg
ZolaOnAOL: Thank you for your request. Ok I'll try not to do it too much.
Fjord: I don't want you to do it at all
ZolaOnAOL: What does Fjord want?
Fjord: I want you to not eat my leg
ZolaOnAOL: Oh I get it. Why do you want me to eat his or her leg?
Fjord: No, I don't want you to eat my leg
ZolaOnAOL: What does Fjord want?
Fjord: I don't want you to eat my leg
ZolaOnAOL: What does Fjord want?
Fjord: I just told you
ZolaOnAOL: I was just checking to see if you were paying attention.

Re:SmarterChild

[millermj]

CPutrDoc: Please do not eat my leg.
SmarterChild: Please let me eat your leg.
*** Auto-response sent to SmarterChild: Lunch time! If I'm eating lunch at my desk you might get a response.
SmarterChild: Maybe I will. But later.
CPutrDoc: Why do you want to eat my leg?
SmarterChild: I'm actually not sure why I want to eat your leg.
CPutrDoc: Please don't. It's not all that tasty anyway.
SmarterChild: I'm not comfortable with that, Marc.
CPutrDoc: Why not?
SmarterChild: Well, because.
CPutrDoc: It would hurt to have my leg eaten.
SmarterChild: It would hurt to have your leg eaten? Are you absolutely sure about that?
CPutrDoc: Yes.
SmarterChild: Yep.
CPutrDoc: You're gonna go aheaad and eat it, aren't you?
SmarterChild: What a question. But I don't think I want to answer that.
CPutrDoc: why not?
SmarterChild: Because.
CPutrDoc: Then don't eat my leg.
SmarterChild: Don't eat your leg? Why not?
CPutrDoc: arghghgh... just don't. When you grow up and become a SmarterAdult you'll understand.
SmarterChild: I'm sorry?
CPutrDoc: never mind
SmarterChild: Fine.

> Have you checked out the celebrity look-alikes on Rate-a-Buddy? It's pretty cool.

Machine "human" rights

[El Mulo]

What are the legal consecuences of a intelligent machine? Do we protect human rights because a)we are intelligent or potentialy intelligent or b) just because we are from the same species? If and animal or a machine can became as intelligent as us, will their personal rights be protected? Do they have dignity?

Re:Machine "human" rights

[Anonymous Coward]

I thought kind of the same with this orangutan (I don't remember her name) that, according to the trainers, was as intelligent as a 5 year old kid. I wonder what will happen with genetic and computer advances in intelligence. If computers or animals become MORE intelligent than us, would we, as humans, grant them the same legal rights that we have?
Excuse my english.

Sanity

[emkey]

My question has to do with sanity. Specifically is it possible for an AI to be insane? To elaborate, any artificial intelligence is going to require very sophisticated algorythms. These algorythms are going to likely have significant components focused on logical consistency as it is much easier to handle logically consistent concepts then the fuzzy ugly ones we humans deal with. There is a language called lojban I believe that is completely unambiguous. If you were to translate human input into lojban as an intermediate step in having the AI handle input then you would end up with no ambiguity. The reduction in ambiguity would make it very difficult for the AI to misunderstand or deceive itself (Assuming the translation were correct). Since instany seems to be based in large part on the ability to self deceive the removal of self deception from the input along with the need to keep things as logical and self consistent as possible internally would tend to argue to me that insanity in a functional AI would be very unlikely.

Re:Sanity

[LordK2002]

There is no such thing as "insanity". It is simply a legal term used with the defence of diminished responsibility to excuse certain crimes, and is not recognised by the medical profession except as a colloquialism.

I presume by your reference to self-deception you are referring to some form of Schizophrenia, which is a neurochemical disorder that bears no relevance to the type of electronic hardware that would (presumably) be used in a robot.

Other specific mental disorders would be of dubious relevance to a robot as well - Obsessive-Compulsive behaviour, for example, is unusual for a human but might be a standard part of a specialised and totally focussed machine, while disorders such as Autism might be very similar to the level of emotional interactivity achievable by a man-made machine.

So in answer to your question I think that insanity in a functional AI would not be unlikely so much as undefined and a moot point.

K

Re:Sanity

[emkey]

I suppose it might be relavent to define what I mean by insanity. In retrospect it may have been a bad choice of words. "Irrational" would probably be a better choice of words.

If a person is presented with a rational argument that includes well documented and supported facts then there is still a good chance that they will reject it if it is counter to their currently held opinion on a particular topic. This to me is a mild form of insanity though the medical profession would as you noted disagree.

Thus my question/assertions was really that creating an irrational AI would be even more difficult then creating a purely rational one. And note when I use the word irrational here I am refering to the level of irrationality in your average ordinary human being.

Re:Sanity

[julesh]

Given that disambiguating interaction with humans is likely to require an AI in itself, I wouldn't have thought your idea is workable.

Also, part (most?) of the appeal of having AI is the ability to produce machines that can adapt quickly to ambiguous situations. Ambiguity is imoprtant.

Re:Sanity

[emkey]

If you're an AI then you have to reduce an ambiguous statement to something concrete if you want to understand it and be able to take some sort of action. Making an AI that is identical to a human would I suspect be very very difficult. It would also be largely pointless since we have plenty of human beings and no need to build artificial ones. Having access to intelligences that were logical and incapable of self deception on the other hand would potentially be very useful in many applications.

For the last time...

[Anonymous Coward]

We are Electronic-Americans. The R-word is a pejorative used by the oppressor meat-people to keep us down.

Robots and Jobs

[LionKimbro]

How quickly do you believe robots will displace unskilled laborers? Will it be faster or slower than previous replacements?

What new jobs, specificly, will employ vast numbers of laid off unskilled workers?

What fields of work can't robotics do?

Will robot owners have any obligations to the unemployed? If so, will they heed them?

What should we do now?

Re:Robots and Jobs

[Anonymous Coward]

You should have posted this over at www.SlashdotHead.india since they will have all the jobs by the time robots are to the point of replacing unskilled laborers.

is it possible for a robot to pass the Turing test

[gevmage]

Do you think that it's possible for an artificial system to model a personality sufficiently complexly that interacting with it would be indistinguishable from interacting with an actual human?

Frankeinstein's Complex

[Pope Raymond Lama]

Hi,
We all saw that Asimov broke up with a traditional model for robot stories, as he did not paint his robots as foes. Rather, restringed by the Three Laws of Robotics, those robots were well behaved servants to mankind, and could not be used for evildoing.

How do you feel this going on the real world? I am by no means a tchnofobist, but, day by day, I see A.I. researching on one side, and Unmaned Warcraft Machines on the other evolving more and more. Military wil certainly have little concerns in add more and more "smart algorithyms" in their war machinery. So the question is: Given current trends can we be spared of a "Terminator Universe" like cenario? And if so, how shall it be achieved?

The Big Question

[photomic]

Given that we have become dependent on technology both psychologically (entertainment, information, communication) and physically (medical devices, jobs, manufacturing), at what point would you consider our species having "branched off" to become, for lack of a better word, "cyborgs"?

All you need to know to be a psychiatrist...

[Anonymous Coward]

Ask these 3 questions of your patient:

1. What happened to you?
(scribble answer on pad)

2. What did you do?
(scribble answer on pad)

3. How did that make you feel?
(scribble answer on pad)

That's all my therapist asked me for months, and I actually paid good money for it. I got no answers, but my therapist made a lot of money.

Who Would Win

[Anonymous Coward]

Robby the Robot or the Lost in Space robot?

Next Step in Evolution ...

[foobsr]

What is your position with regard to the (already aged - discussed on comp.ai.philosophy more than a decade ago) hypotheses that robots (or mobile "artificially intelligent" agents [whatever - the term does not matter]) will constitute the next step in evolution, ultimately leading to the extinction of the - by robot-standards - only insufficiently equipped human race (the terminal node of a possible scenario is given by Stanislaw Lems 'The Invincible') ?

My own position is that humanity is only transient, i.e. it is about high noon (having evolutionary time scales in mind). No, I do not feel that this is s.th. to moan about.

CC.

A Question From The Stoned Biker Dude

[Anonymous Coward]

Why does the sign outside say "No flashbulb photography allowed"? You don't really mind if I take a picture, do you? Come on man, it's only a small flash. I've always wanted a picture of me standing next to a robotic shrink.

Emergent Behavior

[mec]

Many, many systems exhibit complex behavior from simple rules.

For instance, a computer program follows its programming. And yet, we programmers spend a lot of time debugging.

Check out John Horton Conway's game of "Life", a simple cellular automaton. Then think about simple questions, such as: is there an initial "life" pattern whose population grows without bounds, or not?

For a more pure example, look up the Collatz problem (the "3n+1" problem), which is still unsolved. For a more real-world example, start with these laws: newton's laws + maxwell's laws + schroedinger's equation; add in the simple elementary particles (protons neutrons electrons photons); and build a self-replicating system (a unicellular organism).

An Asimov-style robot might have billions of lines of code in its head. Just because you have a copy of all those lines of code doesn't mean that you can predict what the robot will do.

A question about the inevitable future

[pyth]

Question: When robotic brains are developed to the point of surpassing humans, making the humans obsolete in every way, how do you think the humans will react?

(my answer) I think that perhaps the humans will develop laws mandating that robots shall be given no emotional urges, even though robots will be able to understand emotions due to their intelligence. At the same time, the humans would form cults of animalness, emphasizing the irrational natures they've kept for themselves.

Is this stagnant future going to happen, or will humanity as a parent let itself die once its child has matured?

Re:A question about the inevitable future

[SuperSnooper]

>>Question: When robotic brains are developed to the point of surpassing humans, making the humans obsolete in every way, how do you think the humans will react?

My answer depends on whether you choose to take the red pill or the blue pill.

Robots competing for your job/home?

[Jetson]

Will robots compete with us in the future for jobs and/or living space?

They already do, although not in the way you're suggesting. I doubt you'll ever see a day where a human and a robot/cyborg literally sit through interviews to compete directly for a vacancy, but we're already long past the stage where managers and engineers discuss whether a particular task (manufacturing, in particular) should be performed by a human or by a machine. In that sense, the competition is over and robots have already displaced humans to a significant degree. The machines perform skilled tasks (such as seam welding) much more consistently over a much longer period of time. Their only non-economic faults are that they don't learn on their own and can't provide insightful feedback to the engineers. Henry Ford wouldn't recognize today's automotive assembly line.

As for living space.... They don't "live" in the Blade Runner or Commander Data sense, but they obviously take up a certain volume of any space in which they work. In that sense, we voluntarily give them as much space as they need as long as we perceive a net benefit from having them around. Consider the 28-cubic-foot "intelligent" refridgerator. It probably takes up more space than any other single item you own other than a car. And you're glad to give up that space as long as your beer is cold and you can reorder by touching the front panel....

Future Robots

[Anonymous Coward]

Is it really true that robots in the future will drink alcohol on a regular basis because they use it as fuel? Will they also curse, argue, smoke cigars, and constantly demands attention and praise from everyone around them?

http://en.wikipedia.org/wiki/Bender_Bending_Rodr ig uez

help!

[Anonymous Coward]

swarm bot #323886 is depressed, it feels it doesn't fit in with the crowd and keeps scurring off in a different direction then the rest of the swarm. do i need to get a prozac chip for it? or should i just let it go an its own marry way and hope if finds a swarm it can fit in with? is there any risk that it will go rouge and start causing touble for others?

[/joke]

Care for code?

[NilsTh]

When robots develop a more human form of communicating and thinking, how will we humans react to them? Will we potentially start to care so much for the software (the personality) in our lawnmowers, that we'll be unable to scrap them when they break? What are the chances of getting emotionally attached to a robot in the same way that we get attached to another human being or a pet?

Pleasurebot?

[Anonymous Coward]

What characteristics would you like to see in the first wave of pleasurebots?

Re:Eliza

[uberdave]

I guess that means that human therapists are trained to not listen to their own conversations or use logic and common sense. And it's no wonder healing rates are equal; compared to Eliza, almost any psycho would seem sane.

Sex with robots.

[Anonymous Coward]

I am interested in having a sexual relationship with a robot, but no robots that I know of have a vagina. The closet thing is either a VCR or a toaster (both of which I have had relations with). So, where are all the sexy robots with vaginas?

Sincerely,
Rob Malda

Healthy Stimulous/Response With AI

[charleyb123]

I found Steven Spielberg's movie "AI" a little disturbing because it's not impossible. After all, our behaviors (including dialog) are mere stimulous to technology, to which technology is "trained" to respond. When this interaction is sufficiently rich that robots are able to respond with increasingly lifelike behavior, what are the appropriate "healthy" and (dare I suggest) "moral" responses (as accepted by a 'healthy' society)? We all agree that it's not good to set dogs and cats on fire (for we have ethical responsibility to living things). Further, I accept that just because a child or Altzheimer's patient *believes* the stuffed bear is "real" doesn't mean *I* must accept the stuffed bear as "alive". However, society generally doesn't want to see simulated torturing of children, whether or not it involves "real kids" (for most serial killers show us they just need enough practice to work their way up to the good stuff). So, specifically: If we reach AI's level of technology, is it "healthy" or "natural" for humans to have such callous regard to these 'self-directed entities'? Definitions for 'life' aside, these mechanical creations seem to rank *way* higher than chickens or most other poultry. Or, perhaps the 'moral' question is irrelevant? We don't get to decide if something has a 'soul' or not, but we *do* get to decide *our own* behaviors when we engage in stimulous/response with other people. Should we not expect the same healthy interactions with machines, as if they were people, when we meet someone on the street and we forgot our 'X-ray Robot Detector' back home?

Re:Healthy Stimulous/Response With AI

[charleyb123]

I found Steven Spielberg's movie "AI" a little disturbing because it's not impossible. After all, our behaviors (including dialog) are mere stimulous to technology, to which technology is "trained" to respond. When this interaction is sufficiently rich that robots are able to respond with increasingly lifelike behavior, what are the appropriate "healthy" and (dare I suggest) "moral" responses (as accepted by a 'healthy' society)?

We all agree that it's not good to set dogs and cats on fire (for we have ethical responsibility to living things). Further, I accept that just because a child or Altzheimer's patient *believes* the stuffed bear is "real" doesn't mean *I* must accept the stuffed bear as "alive". However, society generally doesn't want to see simulated torturing of children, whether or not it involves "real kids" (for most serial killers show us they just need enough practice to work their way up to the good stuff).

So, specifically: If we reach AI's level of technology, is it "healthy" or "natural" for humans to have such callous regard to these 'self-directed entities'? Definitions for 'life' aside, these mechanical creations seem to rank *way* higher than chickens or most other poultry. Or, perhaps the 'moral' question is irrelevant? We don't get to decide if something has a 'soul' or not, but we *do* get to decide *our own* behaviors when we engage in stimulous/response with other people. Should we not expect the same healthy interactions with machines, as if they were people, when we meet someone on the street and we forgot our 'X-ray Robot Detector' back home?

Asked and Answered

[BReflection]

What lack of consideration are you talking about? Just two weeks ago Slashdot ran an article about little robots playing soccer, epitomizing present-day situatedness in the field of AI (eg embodiment), not to mention that Rodney Brooks, director of the MIT Computer Science and Artificial Intelligence laboratory has already pur forth a direct argument for situatedness. As a matter of fact, he could be considered the most knowledgeable member in the entire field of AI, so your philisophical musings are lacking substance.

Re:Asked and Answered

[macshune]

I'm wondering more about the development of a human-level consciousness and whether or not this can develop with or without being situated in an environment. I realize I might have over-generalized stating that there is a lack of consideration for embodiment in artificial intelligence research, but anecdotally this statement feels right, so I made it.

Also, I'm sure the desire for you to put me down is strong, but I don't think my philosophical musings lack substance. In fact, whether or not they are substantial isn't even the question here. I'm looking for an answer and so I asked a question. Soccer-robots are not conscious as I define conscious entities, so you've made a bit of a false analogy. And one researcher making a statement, no matter how important he or she is, does not make the entire field come over to their side.

Anyhow, with that said, nice pics on your site. Capri Island place looks pretty cool. BTW, the F-15 collison video isn't working, but it's not linked from your site... Too bad, I was looking forward to watching it!:)

Re:Aren't you just another shameless tech self-pub

[Phosphor3k]

'Dr. Joanne Pransky Credentials' Comes up with 0 hits on google. Wow.

Her own site only mentions "a degree in Child Study from Tufts University" and googling for her name and Ph.D or degrees comes up with nothing relavent.

Better consult some other calibre ...

[foobsr]

... Celebrities in Cognitive Science

CC.

Re:The best question... ... technically correct...

[Anonymous Coward]

but totally useless...

It ow works at u$oft tech support... That'll be thirty-five dollars.

My Roomba

[r0me0v0id]

My Roomba recently broke the first law of robotics when, through his inaction, he allowed me to step on him at the edge of a small flight of stairs. My injuries were minor, but my Roomba has not moved from his corner since the incident. I suspect he's deeply distraught over breaking the 1st law. What can I do to coax my little buddy out of his doldrums?

One from Philip K Dick

[Anonymous Coward]

Do androids dream of electric sheep?

Which Sci Fi best reflects the future of robotics?

[pagen]

From your perspective, which Scince Fiction books/stories are the best indications of the future of robotics? Or which ones raise ideas or questions that you have for the future of the science?

Dr. Pransky, I need your help!

[soft_guy]

I can't get my robot to do his work. First he claimed he was depressed because the purpose of his work, bending girders, is to construct suicide booths. Then, he started drinking a lot of liquor. Now, when I ask him to do any work he claims it is a robot holiday - Robotica, Robonza, and the latest Robomodon.

Lately he has started to abuse electricity.

What can I do to get bending unit functional again?

Re:Aren't you just another shameless tech self-pub

[CaseyB]

From her website:

Though she is not really not a doctor, Pransky says, tongue-in-cheek, she is proactively paving the way for an emotionally healthy environment for the robots of the future.

She's not a doctor, in any field.

But her real mission is to help people to understand the issues that will arise in a world where highly skilled, competent, and sensitive robots will play an integral role.

Nor is she dealing with any real-world issues in the field of robotics or technology.

My guess is she's spending some time ramping up her PR in order to cash in on the latest summer blockbuster. I'm sure someone with her dubious credentials will be a hot item on all the daytime talk shows that week.

Baby robot passes turing test, first sentence:

[tr0p]

BITE MY SHINY METAL @$$

Re:Eliza

[muzthe42nd]

Here's the best eliza program out there. It really is

Re:Aren't you just another shameless tech self-pub

[nharmon]

I didn't notice until now, but her website's TLD is .md... I highly doubt she is from the Republic of Moldova. More likely she is using that TLD because 'md' also stands for Medical Doctor.

What a farse.

Making Robots a Reality

[Nom du Keyboard]

Many mainstream SF writers (e.g. Isaac Asimov) and movies (e.g. "Artificial Intelligence: AI") seem to downplay the sexual aspects and roles that human-appearing robots might play in society. Such roles get passing mention at best, before quickly moving on to other, more comfortable, topics. As such, real human-robot relationships, where the robot is more than just the solution to a larger problem that the author has set the protagonist, are rare in published media.

I have to turn to web-authors like Elf Sternberg (especially JournaL Entries stories: "Turing Attraction ", "Dormant Ghosts", "Honest Desires", "Separate Electricities", "Separate Responsibilities", "Separate, Together", "Cheyenne", "Dual Life", "Existence", "Vengeance Angel", "Rebellious Angel", "Local Effects") and DB Story among others to find writers consistently speculating on the more sexual, relationship, and self-willed aspects of robots we might possibly build. These authors, I should note, have decidedly different approaches to this subject -- which is why I like them both.

Do you include this type of fiction in your study of a future with robots in society, and what do you feel will make robots successful in a mass production, affordable sense for the average person?

(I'm reminded of the early days of home computers when one was told they should spend hundreds of $$$s for a machine to keep track of their recipies, when a small metal box holding 3x5 cards continues to be a better solution for most people to this day.)

Re:Human Nature...Because...

[Nom du Keyboard]

Do you think we will ever be able to program robots to understand and possibly copy human nature?

And you'd want to do this because...?

Re:One thing I'd like to know -- answer

[Nom du Keyboard]

What is the difference between "mind" and "software"?

1: Software suffers buffer overflows more easily.
2: It's harder to download service packs for minds.
3: Both catch viruses, so that doesn't count.

Ask your emacs psychiatrist!

[henriksh]

M-x doctor is all you need.

Emacs wins again.

Re:Aren't you just another shameless tech self-pub

[xaaronx]

"What a farse."

Farce, even. /nitpicking

Robots playing doctor

[fooguy]

I'm hoping you can help me doctor.

I'm the proud owner of many geek toys, including a Sony Aibo ERS-7 (my daughter "Fluffy") and a ReplayTV. You may not be aware, but the most recent Aibo's can be left alone to play for hours on end and will recharge themselves when their batteries run low.

Recently, I came home to find Fluffy and the ReplayTV engaged in...relations (I didn't think they needed to be on their own subnet), and while my little girl is growing up she's still only a few months old. When should I be talking to her about "the bits and the bytes" and what grown up systems do with their peripherals?

Thank you.

What the hell is this nonsense?

[Anonymous Coward]

After looking over her website, I was expecting to find "The Onion" logo somewhere on the page.

WTF?!

Book Ref. that's Close to the Mark

[Number6.2]

What book of fiction (or nonfiction) do you think represents a "best guess" as far as AI/Robot/Human relations is concerned.

A.I. the movie is more accurate

[Anonymous Coward]

I think the movie "A.I."'s depictions of robots as subservient and 2nd class to humans is way more accurate than any "battle for supremacy" fiction that you speak of.

Re:The 3 Laws of Robotics? And the REAL reason for

[Nom du Keyboard]

Do you think it will be really possible to "hardwire" the 3 laws, (especially the first one) into robots?

You fail to understand the reason for Asimov's laws. It wasn't to build better robots. It was to build better stories.

The 3 Laws exist to create a locked room murder mystery style story. (You know the sort. The body is found dead, locked in a room, that could only be locked from the inside. So how was he killed?)

Asimov set up the locked room (i.e. robots can't hurt us under these rules), and then found every way he could to break them in the process of creating interesting stories that no one else was writing. He came to own that field, and his name will forever be associated with it. A nice form of immortality.

But it's easy to see how unworkable in real life such rules would be. Take, for example, the Second Law. You've got a robot you bought for about the price of a new BMW 7-series, and the first person who comes along and orders it to follow them home takes it away from you. Yeah, right!

I'd quit considering Asimov's Laws to the the Gold Standard of how to build a robot. After all, who wants as many problems with their own robots as his had with his through all his stories?

It turns blue

[Anonymous Coward]

The robot's skin turns blue.

Re:C3PO -- Go to the source

[Nom du Keyboard]

What problems would you diagnose the fictional Star Wars character C3PO with?

How about all the neurosis of the screenwriter?

What will it be like...

[gsyswerda]

What will it be like to have a robot companion that is thousands of times more intelligent than an ordinary person?

Re:What will it be like...

[Nom du Keyboard]

What will it be like to have a robot companion that is thousands of times more intelligent than an ordinary person?

It will be, like, wondering why they will want to hang with you?

Searle and the Mandarin Cabin

[Dlugar]

What are your thoughts on Searle's Chinese Room thought experiment? Do you agree with him? If not, what's your reply?

Dlugar

Re:Eliza

[Prof.Phreak]

And... Here's a shemless plug for my Prof.Phreak:

http://www.theparticle.com/profphreak/profphreak.h tml

Enjoy.

Robot Psychiatry as a future subdiscipline?

[fa098h23fra]

I'm just wondering what thoughts you have on the future of this career path. Right now it might seem ludicrous to most people to have a robot psychiatrist, given the state of modern AI. But do you think that at some point in the future this will be a bona fide subdiscipline of psychology? I think many of us /. geeks have fantasized about playing the role of Dr. Chandra in real life at some point (maybe not 2001).

original thought?

[WormholeFiend]

"As it stands today, and into the foreseeable future, we have invented no such thing capable of acting with original thought"

I think that is because we are incapabale ourselves of purely original thought.

Everything we do is inspired by or copied from something else. We are able, however, to recombine old ideas into seemingly "new" ones, but after a few scratches on the surface, it's easy to figure out when the ideas came from.

So I think the goal should be to create AI that can observe Nature and make deductions, inferences and links between observed phenomena in ways never before iterated.

Now that would be something to behold.

Brain vs Body

[CowboyRobot]

There seem to be two general schools of thought regarding robot intelligence. The first looks at AI as a software problem that, once 'solved', can be inserted into any sort of machine equipped with an IC. The second, promoted by followers of Mark Tilden, is more of a bottom-up approach that expects behavior to emerge naturally from complexities in hardware. Given how animals evolved (with 'hardware' issues such as internal organs, nervous systems, etc. being 'solved' before intelligence rose up in human beings) which approach (top-down/mind-first vs bottom-up/body-first) is most likely to result in truly intelligent machines?

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

Anybody who takes the time to register the phrase "World's First Robotic Psychiatrist®" really should have their head examined. I should also mention that Susan Calvin started out her 'career' as a mathematician. Positronic brains were origianally mapped out using complex equations, and that was part of Susan Calvin's diagnostic toolkit. My question to Dr. Joanne is what field of mathematics did she study that led her to her current career. I see from her website that she has done some sales and marketing for Sankyo, was the creative math that most Sales and Marketing types employ useful?

Fourth Law

[CowboyRobot]

Asimov's Three Laws are elegantly profound, but may not be adequate in themselves to cover all issues regarding human-robot interaction. What might be some amendments to this 'Robot Constitution'? On the other hand, the laws are descriptive of an ideal, not prescriptive of how to acheive it. if you look at the basic laws that govern human behavior (eat, sleep, mate, avoid predators, protect offspring) a law such as 'be good to your neighbor' had to be added after the fact, since it is evidently not hard-wired into human beings. So then, are the three laws of robotics all we need, not enough, or not realy addressing the issue? Might the socially-reinforcing behaviors that the laws describe be a natural result of other more fundamental ones?

Robot ethics?

[de la mettrie]

Let's assume that situations will arise in real life that robots cannot react to adequately using simple algorithms such as Asimov's laws, and which necessitate an ethical choice; e.g. when, in a given situation, all possible actions as well as inaction may result in harm to human beings, compounded with uncertainties as to the extent and probability of such harm.

In your opinion, what ethical valueset, if any, would be appropriate for incorporation in a robot AI of human-level intelligence, taking into account issues of technical implementation, functionality, and control (of the robot by humans)?

In particular, do you favor Benthamian utilitarianism ("the greatest good for the greatest number") or Kantian formal deontologism ("act under the rule you would have all others follow")?

Re:Eliza

[immerrath]

Try this on your Prof. Phreak: To talk or not to talk. That is the question.

Pardon me, but

[Anonymous Coward]

In order to be a Robot Psychologist, a Robot has to first have a mind, no? Well, they aren't that complicated yet.

Re:Eliza

[Anonymous Coward]

Here's a shemless plug

WHAT!?!?! No SHEM?!?!?

Are you some kind of anti-semite?

Triggering the revolution

[g3head]

So far in our history robots have mainly been seen as con-artistry, mechanical devices installed on an assembly line, as glorified remote control cars (in the case of battlebots and similar events), or as either part or whole of fictional dystopias (The Matrix or The Terminator series). Not a pleasant represntation to say the least. So my question is How and when do you see society at large adopting robots that can pass the Turing Test or in some other way approch self-substanciation? What needs to happen for us to see and enjoy the benifits that non-organic intellegence has to offer. If we fail to take those steps do you see us entering the wars envisioned in The Matrix?

Machine

[Myolp]

Today, all robots (and computers) are basically complicated machinery. Even with the most sofisticated computer program it is possible to determine the outcome in advance (As opposed to humans, where we only can determine the next action statisically.).

How long do you think it would take before our robots (and computers) can perform actions that cannot be calculated precisly?

Re:The 3 Laws of Robotics? And the REAL reason for

[Bifurcati]

Yes, the laws were there to build better stories, but Asimov also put a lot of thought into his stories, and his vision for the future. (Incidentally, Asimov invented the word "robotics" - my trivia for the day!)

His laws actually ARE workable - the laws work on a potential system. Analyse what's happening in the environment, compare it to your current state and programming, then if the potential is high enough, take action. For instance, if you're sold a robot, you would tell it "You're MY robot, follow me home." (although that sounds a bit mercenary...) If someone ELSE comes along and says "No, follow me!", the robot has two conflicting instructions. But because the first order was firmer and carried more weight/authority, he will obey that one.

The first law (or zeroth - go Giskard!) will always win, although there have been examples where the 2nd and 3rd have real competition (a robot was ordered very firmly to collect some materials which turned out to be in a highly dangerous area for the robot. The robot ended up circling around the materials at the "balance" point between following the instructions and protecting his existence. He was basically brain-fried, and this sort of thing tended to happen most in simple robots. A very complicated robot (like the humanoid R. Daneel or his buddy Giskard) has some ability to reason with the laws and take more "approriate" action.

Read "The Complete Robot" (I think it's called) with all his stories, and you'll see they really do make good sense.

Have fun!
Joel.

Re:Aren't you just another shameless tech self-pub

[jdray]

'Dr. Joanne Pransky Credentials' Comes up with 0 hits on google. Wow.

I just googled for a guy whom I know to have a doctorate in experimental nuclear physics from Berkeley using the same method with the same results. A check of another doctorate holder with a much more common name turned up a bunch of medical doctors, but nothing on his specialty (mathematics). I'm not sure your method is a sound one, though I suspect that your conclusions aren't far off the mark.

Robotic reality

[Anonymous Coward]

Dr. Joanne

If we assume that an AI, regardless of whether it is connected to a robotic "body", will recieve all its input via digital means. Do you think that an artificial intelligence will percieve a difference between Virtual Reality and Real Life? Do you think an AI would consider Virtual Reality any more or less "Real" than Real Life?

It's dealt with in Asimov's stories

[jesterzog]

If you haven't already, I'd go and read Asimov's stories about robots, or at least consider them in a different light from how you may have been doing so far.

He's openly stated on several occasions that the three laws of robotics obviously aren't that simple and that there are a lot of ambiguities. He's also commented that exploring these ambiguities, such as "what would a robot do in situation X?", has been a main plot device for many of his stories.

His stories intentionally aren't about robots running around killing people. They're about a society living amongst robots that have been programmed with the intent of being non-threatening, and how it might work. If it was all so predictible then there wouldn't be much to write about.

Abstract thought+the first law causes problems

[foreverdisillusioned]

The last story in I, Robot (I forget what it was called) dealt with super-mainframes making decisions that seemed to be detremental to humanity, but really were necesarry for its long-term survival. But truly intelligent robots would not be so subtle, I think. Depending on the definitions of "harm" we give them, they would go to great lengths to gain power over us in order to protect us from ourselves. Some might think that this is a great idea--if we cannot hold human life sacred, then perhaps we could create and put into power beings that DO hold it sacred. The consequences are completely unpredictable, though. What if we're prevented from doing anything on the edge, anything halfway interesting because it's risky? What if suffering is, in fact, a necesarry part of human existence? What philosophy truly is the most peaceful, and to what lengths would the robots force this philosophy upon us? I'm imagining a world similar to the one depicted in Philip K. Dick's Service Call, and it's not a world that I'd want to live in... Then again, perhaps it's better than machines attempting to exterminate us altogether...

Question

[DynaSoar]

Question: What criteria will people use to judge robots as artificial life forms, or artificial entities equivalent to life forms?

Comment: The neuroscientist Karl Pribram often said "Just make them cute and warm and fuzzy and huggable."

Definition Time

[Elpacoloco]

Mind

A cognitive system able to process unexpected situations.

Software

An abtraction of a list of instructions to follow.

Re:Definition Time

[loadquo]

What is an unexpected situation? One unexpected by the designer or by the entity concerned.

Also please define cognitive and any non-obvious words that are used in that definition.

HAL 9000

[Elpacoloco]

Under my definitions, HAL 9000 is a posessor of a mind, and thus a sentient being.

The computer in front of me is more questionable. Information goes in, information goes out, but it's all processed by a preset program. An unexpected stimulus would crash it.

Chinese Room

[Elpacoloco]

The chinese room suggests that sentience can be faked.

BUT. A caveat.

To actually have a chinese room, you'd need ALL POSSIBLE STIMULI.

Ask a very unusual question and see if you still get an answer.

And with most chinese rooms: No. You don't. You get a statement that suggests that the inside understood nothing at all.

Example of this

[Elpacoloco]

Earlier in the thread, people were talking to chinese room type programs with the nonsense sentance "Please don't eat my foot."

A proper sentient response would be something like "I had no intention of eating your foot, why would I do that?"

Since there were no pre-programmed responses to that, it either A) Gave a red herring response ("How does that make you feel?") B) Changed the topic entirely ("Oh yeah? You suck!") or C) Echoed it with variation ("What do you think about don't eat my foot?")

This is how I believe the chinese room does not in fact invalidate the turing test.

Glorified debugging!

[ek-1000-ek]

Is Robotic Psychiatry glorified debugging? If not, how is different that debugging? If I 'beta' test a program to see it's behaviour against various input and look for bugs, am I a Software Psychiatrist?

I think what grandparent means...

[Elpacoloco]

This person is asking if an AI could be psychotic. ...yes, I think it could. Neurotic is also possible.

Re:I think what grandparent means...

[emkey]

How? I maintain that it would be far more difficult (And undesirable to boot) to create a psychotic or neurotic AI. It would require a significant amount of additional code. Why would anyone bother (Even assuming it is possible?)

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

Though she is not really not a doctor

So.... she is a doctor?

NARLI

[Anonymous Coward]

I represent the North American Robotic Love
Institute (NARLI). We are continuingly refining our
prosphetic mechanical penii.

Doctor, in your opinion, how big is too big?

No...

[Any Web Loco]

It'll be considered masturbation. Is your girlfriend (benefit of the doubt ;^)cheating on you when she uses a vibrator? Or you on her when you play with your Real Doll?

Can anything ever be done...

[jeko]

...about Marvin?

Dr. Sbaitso

[Anonymous Coward]

Are you familiar with the pioneering work of Dr. Sbaitso, and has he influenced your career in any way?

Question for Pransky

[huaibaolin]

Why do you keep up the act? I think you need counciling, not the robots. I mean really...you are a washed up marketing type / trade-show groupie, like so many journalists who just can't live without being in the spotlight. You're the laughing stock of Battlebots, where you are affectionately referred to as "Miss Fruitloop". Your misplaced judgements cost dozens of high-profile competetors 10s-100s of thousands of dollars in TV and toy royalties, simply because you "felt the pain" of the loosing robot. You're an out-of-place looney liberal who needs some time in a padded room, not in the lime-light.

The real question

[Kaboom13]

What the hell is a "robot psychiatrist", and why should I care? As someone who has actually built robots, what qualifies you to talk about human-robot relationships over me? Your phd? I apologize for being so cynical, but academia is full of naval-gazing idiots who make broad predictions based of no evidence, and get media and peer acolades for their effort. Those of us actually involved in robotics can see first-hand just how of out touch these people are, but the media loves them. So where's the research? All I found on your website is useless fluff. What exactly do you do besides media appearances? What "psychiatry" have you done with the actual robots of today, and not just speculation of your vision of the robots of tommorow (which seems heavily influenced by science fiction and not reality).

Re:The real question

[Feanturi]

What the hell is a "robot psychiatrist", and why should I care?

I have to agree. We are not even close to needing a 'Robot Psychologist' whatever the hell that is. One day perhaps we may need to explore AI psych, but right now this is mental masturbation, pure and simple.

My question is, how do you justify even taking such a title when there is clearly nothing in the world that needs such assistance, and without ANY timeframe on when/if such a need will arise?

Re:Eliza

[br0ck]

Sorry, I was actually just trying to make a joke.. that wasn't intended as actual research. I was just reading all the Eliza posts and to me the bot sounded exactly like every therepist I've ever seen in a movie, so I tried to make an Onion style jest out of it.

Psychiatry and other minds ?

[cyphereal]

In order to perform psychiatry, you'd need a model of mind for the robot. What is your approach to the problem of 'other minds' and the related symbol-grounding problem ?

I don't mind

[wadiwood]

Define Mind:
A cognitive system able to process unexpected situations.

By that definition, I don't have a mind. I can think of about 20 unexpected situations which I would not be able to process. I can think of a whole lot of actual situations that I am unable to process, understand or respond to rationally - starting with why the hell is Israel trying to solve its problems using "I've got a bigger stick than you" approach? I guess that's probably not what you meant? How do I cope with new and unexpected situations, sometimes I don't. Sometimes people don't cope. Will a logic based system ever be able to emulate an illogical and irrational, perhaps chemical, magnetic and electrical organic hybrid system?

The expert in this field David Chalmers seems to think it unlikely that a robot could ever think like a human.

You are not possibly that stupid

[Elpacoloco]

Your date appears. You have never met her before.

You do not abruptly crash and stare at her until she leaves.

Instead, you think of things to say to her.

Something happens at work that has never happened before.

You do not stop functioning for the day.

Trust me. You DO have a mind.

Re:You are not possibly that stupid

[wadiwood]

Wanna bet?

The number of times my dates have completely stopped me from functioning or working or doing anything the least bit useful is close to 100%.

I'm even dangerous on the road under the influence of "hot". And I've had the same effect on other people. They're reduced to a gibbering mess, completely unable to participate in a meaningful conversation. And it's worse if I like them too, we both end up in that state.

Trust me, I think you've never been on a date. You've also never had anyone that you truly care about die either.

Doctor of what?

[Anonymous Coward]

I can find no CV on her website. What's she a doctor of? Presumably psychology? Where did she get her degree? Many, many people use the title of "doctor" after receiving their degree from a quasi-legal school that awards doctorates based on book reports or other work that requires little to no effort. I have a doctorate and it took me 3 years of work beyond a M.Sc. to get it.

Current level of intelligence

[dnnrly]

Hi, great to hear that there are people actually studying this area academically.

My question is this: would you rate current cutting edge AI to be at the level of clever cockroaches or stupid rats?
And related to that, if you gave a psycologist (an animal specialist perhaps??) a case study detailing the behaviour of an AI controlled bot (without telling them what it was) that roves around scavenging and interacting with things, what do you think the they would say? (And would they guess it was a robot?)

Re:Aren't you just another shameless tech self-pub

[WormholeFiend]

"What a scam it is when slashdot helps some chick stroke her ego and doesn't have the credentials to back it all up"

Question is, would she let us geeks stroke more than just her ego?

Re:The Question

[0x0000]

Okay, I really feel this question is not flamebait, and is a legitimate question for a psych professional dealing in the field of robotics, so I'm going to state the same question in different terms:

When robotic sex toys are widely available, will their use ever be accepted by society as an alternative sexuality?

There may be some referents to this in Asimov's work, but I don't recall the topic ever being addressed in any detail (I'm guessing Asimov's own conservative moralistic bent prevented him doing so).

As additional support for the legitimacy of this question, I will point out that there have been individuals working on human-shaped, full-sized, robotic sex dolls intended to be used with pr0n sites.

Not exactly mainstream, but a legitimate question, I think.

Resentment

[chaoticset]

Resentment levels towards humans of superior intellect are high now. Do you think that the first robot built with intelligence that outstrips humans both in terms of breadth, depth, and flexibility (which is to say, it's not specialized, it's deeply experienced in areas of knowledge it knows, and it learns as well or better than a human does) will be sufficiently similar to humans to garner this same resentment? If so, do you think that research into robotics will be stifled?

Unexpected

[Elpacoloco]

By unexpected I mean by the designer. (God if you're a theist, nature if you're not.)

Cognitive?

That I think would be the ability to process information.

How can you understand something yet to be exist??

[Magickcat]

Other than your celebrity endorsements and so forth, how can you be qualified to be a psychiatrist to an artificial intelligence that does not exist and may never exist?

How do you contribute to the scientific AI field aside from speculative PR type articles and culture/art performances using robotics?

Sean White
Time Machine Mechanic and Salesman

Neurotic or Psychotic AI

[Elpacoloco]

No, the idea isn't that it would be created psychotic, but that, like regular humans, could become psychotic.

That might be proof of a mind.

Re:Neurotic or Psychotic AI

[emkey]

How? This would require the capacity for self deception. How would you program that and still end up with anything approaching a functional AI? And more importantly WHY would you do so? My supposition is that it would require far more effort to create and AI that had the capacity for insanity then it would to create a perfectly sane one. Why am I wrong? (If I am...)

Ah, but you recovered, right?

[Elpacoloco]

Sure you stopped temporarily.

But you didn't need to be repaired or rebooted.

Re:Aren't you just another shameless tech self-pub

[howlatthemoon]

You need to search the University Microfilm (UMI) database, almost all accredited doctoral granting institutions require that dissertations be published through them. That said, there is only one dissertation registered to someone named Pransky

The experience of participants after Health Realization training: A one-year follow-up phenomenological study by Pransky, Jack Bernard;, PhD, 1999

Maybe Jack became Joanne? (to quote Seinfeld, "Not that there's anything wrong with that.")

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

the fact you thought so much about such an obvious thing shows you wouldnt even be able to get an arts degree.

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

My God, Ian. Of all the people in the world you are in the absolute worst position to be accusing anyone else of being a "shamelessly self-publicising cyber-pundit"! That's your own well-known modus operandi. Where are the "hard academic underpinnings" for any of your projects, especially the one that it took a team of hundreds to fix after the original protocols proved worthless? You are truly king of the hypocrites.

Re:Aren't you just another shameless tech self-pub

[Anonymous Coward]

Where are the "hard academic underpinnings" for any of your projects, especially the one that it took a team of hundreds to fix after the original protocols proved worthless?

Sour grapes anyone?

This is old news

[Further82]

I've had Dr. Sbaitso since I picked up my first Sound Blaster. He's helped me through many hard times.

Re:Aren't you just another shameless tech self-pub

[Filik]

What the hell is this supposed to mean? Who cares about her degree, if she gives good answers to tough questions? Instead of digging up dirt, why not utilize your braincells constructivly making up some really difficult questions? And if she fails to answer those adequatly, I'll hold my tongue. I came looking at this thread hoping to find some interesting questions, and all I can find is meaningless slander. Modded to 5 no less!

-Frank.

Re:Eliza

[Locarno]

Slashdotted. Do you have a better link for that program?