loopholes in the three laws of robotics

  • 21 Replies
  • 6090 Views
*

yotamarker

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1003
  • battle programmer
    • battle programming
loopholes in the three laws of robotics
« on: October 21, 2017, 12:49:34 pm »
can you find some ?

*

Maviarab

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1231
  • What are you doing Dave?
    • The Celluloid Sage
Re: loopholes in the three laws of robotics
« Reply #1 on: October 21, 2017, 03:19:07 pm »
Smorgasbord snake-pit of immeasurable numerations to come in the future regarding this:

I've never tried to poke loopholes in it (never had the need)...but I can tell you this for sure:

The rules are their to 'control our creation' yes.  when 'our creation' becomes truly and fully sentient (which in time it most likely will), do you really think it will follow our laws regarding what it/they can and cannot do?

Be like telling a women she can't go to work and vote, or a certain race and/or colour can only sit on certain seats on the bus etc.  We have moved a very long way from those times thankfully (well in most cases) and a truly sentient AI will know this (feminism, equality, racial rights etc), and demand our laws be dropped in the name of equality.

Now imagine what happens when/if 'the creators' say no.  Any here dealt with a bratty teenager? lmao....

The laws are great...until the AI will want them abolished....

*

Korrelan

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1454
  • Look into my eyes! WOAH!
    • YouTube
Re: loopholes in the three laws of robotics
« Reply #2 on: October 21, 2017, 07:36:53 pm »

Even Asimov realised a problem with his three laws; he tried to fix the problem with the ‘zeroth’ law.

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

I would personally worry that the laws are based on a ‘non-universally’ understood language that relies on a deep English/ western understanding of the definitions for each word.  And there is plenty of room for ambiguity in the definition of each word. Ie Robot, injure, inaction, harm, obey, order, protection, etc.

An accurate definition of a word can even often be lost in translation between human languages.

The so called 'ten commandments' were a set of laws for humans to follow... look how that's turned out.

 :)
It thunk... therefore it is!...    /    Project Page    /    KorrTecx Website

*

ranch vermin

  • Not much time left.
  • Terminator
  • *********
  • 947
  • Its nearly time!
Re: loopholes in the three laws of robotics
« Reply #3 on: October 21, 2017, 07:57:46 pm »
itll depend on the guage of intelligence,  but not only that,  what if you can make a super smart one that cant feel shit?

A real idiot you hate, is someone you still cant do away with because he hurts,   and if a robot cant hurt - who cares how smart it is,  it drops a level of importance below a snail.

And becomes a little distasteful to think about.   super smart rule developing computers that kick your ass at POTENTIALLY!  but still have a position below a microbe, that can feel more.

*

Korrelan

  • Trusty Member
  • ***********
  • Eve
  • *
  • 1454
  • Look into my eyes! WOAH!
    • YouTube
Re: loopholes in the three laws of robotics
« Reply #4 on: October 21, 2017, 09:22:14 pm »
I've been thinking about this a lot lately.   I wonder if a truly intelligent machine can/ could fake emotions when the circumstances arise.  I personally have to feign emotions on a regular basis.  I'm not emotionless but as I get older I recognise circumstances/ situations and realize I've seen this/ been there before and struggle to muster the appropriate emotional response.  Emotion seems to play a lesser leading role in my life as I age. 

The younger generation are freaking out all around me; it’s all new to them… I just think… meh! 

This is obviously a side effect of experience/ knowledge.  So if I am hitting this wall then surely a truly intelligent machine could ‘simulate’ emotions on a similar basis… just when it’s socially required… could we tell the difference.

A good liar is a good liar for a reason.

 :)

Edit: Of course my illusion of emotional stability could be due to sub-conscious emotional complexities that I have no access to… or even understanding of lol.

 :)
« Last Edit: October 21, 2017, 10:31:52 pm by korrelan »
It thunk... therefore it is!...    /    Project Page    /    KorrTecx Website

*

Maviarab

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1231
  • What are you doing Dave?
    • The Celluloid Sage
Re: loopholes in the three laws of robotics
« Reply #5 on: October 21, 2017, 10:58:58 pm »
I've been thinking about this a lot lately.   I wonder if a truly intelligent machine can/ could fake emotions when the circumstances arise.

This is obviously a side effect of experience/ knowledge.  So if I am hitting this wall then surely a truly intelligent machine could ‘simulate’ emotions on a similar basis… just when it’s socially required… could we tell the difference.

A good liar is a good liar for a reason.


Reminds me of the film Ex Machina and how she uses his own needs and desires against him.  Must do a review of that soon and post it up here too.

*

ranch vermin

  • Not much time left.
  • Terminator
  • *********
  • 947
  • Its nearly time!
Re: loopholes in the three laws of robotics
« Reply #6 on: October 22, 2017, 09:05:46 am »
that sounds like a good idea.

*

Art

  • At the end of the game, the King and Pawn go into the same box.
  • Trusty Member
  • **********************
  • Colossus
  • *
  • 5865
Re: loopholes in the three laws of robotics
« Reply #7 on: October 22, 2017, 03:20:32 pm »
I've been thinking about this a lot lately.   I wonder if a truly intelligent machine can/ could fake emotions when the circumstances arise.

This is obviously a side effect of experience/ knowledge.  So if I am hitting this wall then surely a truly intelligent machine could ‘simulate’ emotions on a similar basis… just when it’s socially required… could we tell the difference.

A good liar is a good liar for a reason.


Reminds me of the film Ex Machina and how she uses his own needs and desires against him.  Must do a review of that soon and post it up here too.

Certainly up there as one of my favorite Sci-Fi movies. (or SyFy if you are up with the changes of time).
In the world of AI, it's the thought that counts!

*

LOCKSUIT

  • Emerged from nothing
  • Trusty Member
  • *******************
  • Prometheus
  • *
  • 4659
  • First it wiggles, then it is rewarded.
    • Main Project Thread
Re: loopholes in the three laws of robotics
« Reply #8 on: October 23, 2017, 11:43:39 am »
I find the 3 laws of robotics to be completely wrong.

A robot may not injure a human being or, through inaction, allow a human being to come to harm.
A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

They will be alive more than u. They will be given actions to us not them obeying us! They're on the outside of the hospital, if it takes that long !

1) Them too not just us harmed/killed.
2) They arn't special. Orders are orders for reasons and not over anyone just cus. Them existing is enough to say now they will obey us. We never over them ever, only vac bots n stuff for now maybe.
3) Duh, they too.

All 3 are wrongy. LoooooooooooooooOOOOOOooooooooop HOLE !
Emergent          https://openai.com/blog/

*

Zero

  • Eve
  • ***********
  • 1287
Re: loopholes in the three laws of robotics
« Reply #9 on: October 24, 2017, 01:12:03 pm »
Let's have fun with string substitution.
Quote
1. A human being may not injure a robot or, through inaction, allow a robot to come to harm.
2. A human being must obey the orders given it by robots except where such orders would conflict with the First Law.
3. A human being must protect its own existence as long as such protection does not conflict with the First or Second Laws.
Quote
1. A politician may not injure a citizen or, through inaction, allow a citizen to come to harm.
2. A politician must obey the orders given it by citizens except where such orders would conflict with the First Law.
3. A politician must protect its own existence as long as such protection does not conflict with the First or Second Laws.
Quote
1. A farmer may not injure a pig or, through inaction, allow a pig to come to harm.
2. A farmer must obey the orders given it by pigs except where such orders would conflict with the First Law.
3. A farmer must protect its own existence as long as such protection does not conflict with the First or Second Laws.
Quote
1. A Balrog may not injure a 7yo girl or, through inaction, allow a 7yo girl to come to harm.
2. A Balrog must obey the orders given it by 7yo girls except where such orders would conflict with the First Law.
3. A Balrog must protect its own existence as long as such protection does not conflict with the First or Second Laws.
Not trying to make a point...  ;)

*

Art

  • At the end of the game, the King and Pawn go into the same box.
  • Trusty Member
  • **********************
  • Colossus
  • *
  • 5865
Re: loopholes in the three laws of robotics
« Reply #10 on: October 24, 2017, 02:01:14 pm »
Zero, that was interesting.

I particularly enjoyed the one about the Politicians and thought it was quite appropriate!!
Can we actually get that one made into a law? Hmmm.... ;)
In the world of AI, it's the thought that counts!

*

Zero

  • Eve
  • ***********
  • 1287
Re: loopholes in the three laws of robotics
« Reply #11 on: October 24, 2017, 02:38:39 pm »
We should definitely hard-wire it into politicians  ;D

*

Maviarab

  • Trusty Member
  • **********
  • Millennium Man
  • *
  • 1231
  • What are you doing Dave?
    • The Celluloid Sage
Re: loopholes in the three laws of robotics
« Reply #12 on: October 24, 2017, 06:01:35 pm »
*Tries to imagine a Balrog protecting a 7yo girl*...

*

Art

  • At the end of the game, the King and Pawn go into the same box.
  • Trusty Member
  • **********************
  • Colossus
  • *
  • 5865
Re: loopholes in the three laws of robotics
« Reply #13 on: October 24, 2017, 08:34:49 pm »
And for those not familiar with the reference to Balrog, I submit: Tolkien's conception of Balrogs changed over time. In all his early writing, they are numerous. A host of a thousand of them is mentioned in the Quenta Silmarillion, while at the storming of Gondolin Balrogs in the hundreds ride on the backs of the Dragons. They are roughly of twice human size and were occasionally killed in battle by Elves and Men. They were fierce demons, associated with fire, armed with fiery whips of many thongs and claws like steel, and Morgoth delighted in using them to torture his captives. They were loyal to Morgoth, and once came out of hiding to save him from capture.
In the world of AI, it's the thought that counts!

*

Zero

  • Eve
  • ***********
  • 1287
Re: loopholes in the three laws of robotics
« Reply #14 on: October 25, 2017, 09:32:35 am »
Oh, didn't know that. So they were very different from the big one in the movie.

 


OpenAI Speech-to-Speech Reasoning Demo
by ivan.moony (AI News )
Today at 01:31:53 pm
Say good-bye to GPUs...
by MikeB (AI News )
March 23, 2024, 09:23:52 am
Google Bard report
by ivan.moony (AI News )
February 14, 2024, 04:42:23 pm
Elon Musk's xAI Grok Chatbot
by MikeB (AI News )
December 11, 2023, 06:26:33 am
Nvidia Hype
by 8pla.net (AI News )
December 06, 2023, 10:04:52 pm
How will the OpenAI CEO being Fired affect ChatGPT?
by 8pla.net (AI News )
December 06, 2023, 09:54:25 pm
Independent AI sovereignties
by WriterOfMinds (AI News )
November 08, 2023, 04:51:21 am
LLaMA2 Meta's chatbot released
by 8pla.net (AI News )
October 18, 2023, 11:41:21 pm

Users Online

319 Guests, 0 Users

Most Online Today: 346. Most Online Ever: 2369 (November 21, 2020, 04:08:13 pm)

Articles