Take the 2-minute tour ×
Science Fiction & Fantasy Stack Exchange is a question and answer site for science fiction and fantasy enthusiasts. It's 100% free, no registration required.

I am currently reading Asimov's The Robot Series. I have read The Caves of Steel and now I'm reading The Naked Sun.

In these two books, Elijah orders (on a couple of different occasions) R. Daneel to do something (not really important what that is). The latter refuses, because that would bring the former to harm, thus contradicting the Second Law.

Then Elijah thinks that he could point a blaster at his own head and claim that he was going to kill himself if Daneel didn't execute that order, but doesn't.

A human, (could - would) doubt something like that if Elijah had said it.

So my question is this:

Would/Could a robot doubt the saying of a human?

share|improve this question
1  
You should start off with the I, Robot series. It goes into interactions between humans and robots that are related to your question. –  Chahk May 23 at 18:33
4  
Better yet, see this answer for Asimov's own suggested reading order. –  Chahk May 23 at 18:42
    
When I input wrong password for my SE account, a robot barks - Do you think I am an asshole? So, yes, they can ofcourse. In Fact, they do. –  Awal Garg May 24 at 5:55
    
@AwalGarg That's a simple check, not a doubt. A doubt would be if you entered a wrong password and then told the Computer 'Dude, it's me; let me in' –  Shevliaskovic May 24 at 7:58
    
hahaha @Shevliaskovic I will try that sometime and tell you what he said.... But most probably, he will say, "Now I think you are an asshole!" –  Awal Garg May 24 at 8:00

1 Answer 1

up vote 16 down vote accepted

It would depend on the complexity of the robot concerned, what the human was saying, and the context.

  • Cutie did doubt what humans told him: he doubted that the two humans made him even though they told him they had assembled him; he doubted that the dot of light they showed him in the telescope was a planet; he doubted everything that Powell and Donovan told him.

  • Herbie would simply know you're lying.

  • Giskard Reventlov would, like Herbie, be able to detect when you're lying.

  • Robot AL-76 would simply accept whatever you said to it on face value - it's too specialised to interpret human behaviour.

  • When Gloria told Robbie, "Bad boy! I'll spank you!", Robbie "cowered, holding his hands over his face". However, later, when she threatened to cry if he didn't give her a ride on his shoulders, he "paid scant attention to this dreadful possibility". He seems to know when she's being serious and when she's just putting it on.

  • And, as for Daneel, he would analyse the situation, assess the likelihood of a human actually killing himself with a blaster, decide it was very unlikely - but still try to talk the human out of the act, just in case.

Yes, some robots could absolutely doubt what a human was saying.

share|improve this answer
    
One important reminder: Unless there is other reason of greatest importance, even if the robot realices the lie, it would not confront the human due to the possibility of harming the human self-steem. –  SJuan76 May 23 at 20:59
    
Excellent answer! You said everything I was coming to say, but added more and did so more eloquently than I would have. Cutie, as you said, is by far the most egregious example of a robot doubting what a human says; he even refuses to believe the books in the station's library, and founds his own religion. –  James Sheridan May 23 at 23:31
    
One more thought related to this: even if the robot deduces that the human is almost certainly lying, it would be very difficult to be completely certain, and even the slightest possibility of a human coming to harm would have to be reacted to as long as it did not conflict with more likely contraventions of the other laws. So, depending on the situation, a robot might have to procede under the assumption that the human will harm themselves anyway (although not in the situations described, where the robot has already identified potential harm from following the human's instructions). –  Jules May 24 at 5:31
    
@SJuan76 unless it calculated more serious harm to the human as a result of not acting on the knowledge that the human was untruthful. –  jwenting May 24 at 8:47

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.