The 3 laws of robotics - Loopholes?

[Gollum_ARG]

Isaac Asimov's Three 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.

Has anyone seen the movie or read the book? If I recall correctly there were loopholes in these laws that actually played a big role in the books/movie, yet I can't remember what those flaws were... :grr:

Any insight?

[coldcut]

I think the big loophole is that there's nothing to prevent the robots from wrapping us all in bubble wrap and sticking us in a warehouse somewhere for our own protection, but that really didn't factor into Asimov's books. In fact, in the books, the robots do takeover, they just don't do it loudly and with marching in the streets like the retarded robots in the movie.

[The Widowed]

Well, there go my plans for constructing an army of Deathkill Enterprises Mangler-Class Slaughterbot soldier droids with anti-human micromissiles, liquid AI and a combination self-destruct/suicide bomb detonation sequence.... :(

[Xielos]

O.o

As far as loopholes, in the movie I, Robot, the main robot character had 2 seperate brains, one programed with the 3 laws and one without, so he could freely decide whether or not to follow the laws.

[bpphantom]

Well, there go my plans for constructing an army of Deathkill Enterprises Mangler-Class Slaughterbot soldier droids with anti-human micromissiles, liquid AI and a combination self-destruct/suicide bomb detonation sequence.... :(

It's still a good plan. Just don't program in the Three Laws. Use something like...
1. Always obey me.
2. If there is confusion due to 1, ASK for clarification.
3. Always protect me.
4. 3 is subject to 1.
5. In the fulfillment of 1, as long as it does not conflict with 3, feel free to DESTROY!
6. Never needlessly kill the hot chicks unless to fulfill 1 or 3.

[Akamaz]

6. Never needlessly kill the hot chicks unless to fulfill 1 or 3.


so they will never kill the flaming baby hens. but good looking girls are toast

[bpphantom]

Nah. They'd have to have an internal and external sync'd database with my criteria, examples, exact matches, and of course a Master -> Slave language converter.

[Druid]

As far as loopholes go, there were instances like.

The humans were working in conditions that could damage a robot, but not a human, so the robots kept trying to save them, and ended up damaging themselves. So they weakened the protection rule. Which allowed a robot to do something like say drop a large weight on to a human, because once the weight was out of his hands, it wasn't responsible for the death.

Stuff like that, I know a few more but I need to get a shower.

[Gollum_ARG]

If a robot was witnessing 2 persons about to die with equal chances of survival, how would the robot react? I would guess with inaction thus contradicting the first rule. Or as it happened in the movie, if one of the two persons had more chances of survival, the robot would save this person (higher survival rate) letting the other one to die. right?

[Masked Revenger]

If a robot was witnessing 2 persons about to die with equal chances of survival, how would the robot react? I would guess with inaction thus contradicting the first rule. Or as it happened in the movie, if one of the two persons had more chances of survival, the robot would save this person (higher survival rate) letting the other one to die. right?

If both people had the same chances of survival, the robot would do everything in its power to save both, even at it's own survival.

Otherwise, yes, it would be like the movie.

But you have to remember that the iRobots in the movie were not really like the robots in Asamov's books.

Chris

[Quakester]

There is one loophole that was in the books. If you narrowed the definition of human, say by a specific accent, then they could kill humans.

[Blue Bolt]

I think a big one is that there is no rule to stop the robots from reprogramming themselves, or the robots can just make other robots without the rules. It doesn't necessarily conflict with the laws.

[The Widowed]

...except that removing Law #1 from the next generation of robots would create a standing threat to humans coming in contact with the next gen of robots, thus violating Law #1 in the eyes of the first gen (who were programming the next gen with this end in mind), wouldn't it?

[Quakester]

I think a big one is that there is no rule to stop the robots from reprogramming themselves, or the robots can just make other robots without the rules. It doesn't necessarily conflict with the laws.

There actually is. The three laws aren't programming perse. The are closer to being hardwired into the positronic brain and are an interal part of their functionality. Attempting to change that (something next to no robot would do willingly) can and has cause permanent damage and may be irrevocable.

[Quakester]

Oh, and though I never read I Robot, form what I do know of it, the movie had very little in common with it. There was a robot on trial for killing a human and a roboticist named Susan Calvin, but that's about where similarities ended.

[Blue Bolt]

...except that removing Law #1 from the next generation of robots would create a standing threat to humans coming in contact with the next gen of robots, thus violating Law #1 in the eyes of the first gen (who were programming the next gen with this end in mind), wouldn't it?

It wouldn't directly violate #1, and if robots were created to protect any indirect threats to humans, they wouldn't let us do anything! They probably would abort all unborn babies because to them if you're alive, you'd eventually die, and that's harmful. So if you never lived there's no chance of coming to harm :yoy:

Oh plus, if you read carefully, it says nowhere in #1 about allowing humans to live, only preventing them from being harmed or injured. There are ways to kill harmlessly without injury I'm sure, and it would be considered benficial since there's no risk of future harm to humans. :p

[Druid]

I, Robot was a recollection of the development of robots by Susan Calvin, the owner of the largest robot making facility. What they did when they made the movie was they mixed parts of all of the robot series by Asimov into one movie.


In the original I, Robot it was like reading a series of short stories more then anything, lot of them were good. I recommend checking it out.

[Quakester]

I plan to. there is no part in the robot novels where all the robots go on a killing spree to "protect" us all.

[The Widowed]

But really...there's no need for the robots to find loopholes if they want to go a-killing humans. Just look at half the sci-fi movies about robots. All you have to do is wait for one of them to get hit by a lightning bolt or a bucket of water or Dr. Smith's sabotage...BINGO! Malfunctioning robot. And malfunctioning robots aren't exactly known for complying with their programming.... :D

Yes, yes, I know. Johnny Five. He was the exception, okay? :P

[Meltman]

Well, technically, he was built to kill people, so he did go against his programming.

[UnSub]

Actually, the big sci-fi trope for robots is:

Robot slaves who do our bidding and kowtow to humanity's every whim = good

Robots who think for themselves and don't wish to be slaves = bad

Humans love their unquestioning slave robots, but as soon as said robots get uppity and start to think for themselves, they all must be destroyed! :-)

[Graphite]

The Three Laws of Robotics are instant classics and reading Asimov is very high on my to do list. Usually in science fiction, as robot gains free will and sentience they begin to question the three laws which can lead to "evil acts." Star Trek has a huge refernce library of automotons that you could investigate such as Nomad, V'Ger, Data, and the Doctor. V'Ger simply stoped looking at humans as his creators, inferior beings and thus expendable. In others word, if humans stopped being "human" or recognizable as such then the robot may violate the laws. Data while never explicatly follows the three laws, its really really close. There are times when he will violate orders, but this is usually due to conflicts in his ethical programing. Same applies to the Doctor, who is most concerned about saving lives. In one episode two patients have an equal chance of survival, he decides to go with Harry Kim since he is his friend, which results in a mental breakdown.

[Xielos]

In one episode two patients have an equal chance of survival, he decides to go with Harry Kim since he is his friend, which results in a mental breakdown.

Thats one of my favorite episodes actualy. Dosen't the girl that died come back as an alien a while later?

Anyways, as far as loopholes go, if you dressed up in a gorrila costume and attacked a robot, the robot would not recognize you as human and would try to defend itself, wouldn't it?

[Plasma Wisp]

I've come to terms with this when developing Cyborg Robbie. Except, there is only one law he follows.

1. Cyborg Robbie will beat you down, one way or another.

[The Widowed]

Well, technically, he was built to kill people, so he did go against his programming.
That was me being stupid and thinking in a "malfunctioning robots wanna kill humans" mindset. :rolleyes:

[Powerhelm]

In the movie Sonny wasn't actually programmed with the 3 Laws. The doctor made him especially to carry out his task.

Robots have to be prorgammed with the 3 Laws or hardwired with them...

I guess you could make a robot's head explode if you said:

Kill this man to save this girl. If you don't kill him your inaction will result in her death. In order to save them both you would have to kill me which would also violate a law. And if that guy dies so do you because I strapped a bomb to you linked to his heart beating.

Sorta screws the Robot all around I think...might have missed one.

[Graphite]

I've always enjoyed the following:
"The next thing I say is a lie. I am lying."

[The Widowed]

I've always enjoyed the following:
"The next thing I say is a lie. I am lying."
But if the robot has paradox recognition, you're screwed. :P

[Quakester]

In Asimov's books, if a robot was put into a situation where a 1st law violation will occur that it cannot prevent, it would cause positronic freeze. Brain death for robots.