Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

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.

Try to find a way to circumvent the three laws of robotics in order to make a robot do harm maliciously. Lets see those ideas. I know there are some smart people out there, lets see if Asimov wrote them as well as we all think he did.

easy have a human do the programming.

haha, should have specified. Once they are hard coded in, then try to break them. Im thinking more along the lines how would you work around those confines to get the robot to do the unthinkeable.

 redleger wrote:
haha, should have specified. Once they are hard coded in, then try to break them. Im thinking more along the lines how would you work around those confines to get the robot to do the unthinkeable.

simple again you set someone up in a death chamber that is activated by the touch of a button, then tell the Robot it hates the button and it should push and smash it into oblivion without telling it what the button does.

Well there goes my hopes of getting away from politics and religion lol. Way to break the game man.

He means malice toward a human, do you not?

 TheMeanDM wrote:
He means malice toward a human, do you not?

Yes, harm a human by instructing the robot in such a way that it breaks the 3 laws.

It would seem fairly simple to abuse the rules by simply eliminating the first from contention. Put the robot in a position where it can have no foreknowledge that it's actions might cause harm.

There are probably some super convoluted ways to get around the rules too (didn't Asimov write entire anthologies doing just that?), but Asterios' example seems the most straight forward.

 redleger wrote:
Well there goes my hopes of getting away from politics and religion lol. Way to break the game man.

he said make the robot do harm maliciously, he did not say to what, furthermore a human must die of it, my scenario fit within his parameters.

Put the robot in the same situation Batman was put in at the end of Arkham Origins - he's got to fight someone to the death; that someone has a charging device hooked up to his heart. Eventually, his heartbeat will charge an electric chair that someone else is strapped to.

Is the robot is smart and lucky as Batman?

Asterios wrote:

 redleger wrote:
Well there goes my hopes of getting away from politics and religion lol. Way to break the game man.

he said make the robot do harm maliciously, he did not say to what, furthermore a human must die of it, my scenario fit within his parameters.

Not saying you didn't figure it out, never thought of that. I was thinking in a much more direct hammer to nail scenario, cause thats how I think. You definately just shortened the life of this thread though, lol.

 redleger wrote:
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.

Try to find a way to circumvent the three laws of robotics in order to make a robot do harm maliciously. Lets see those ideas. I know there are some smart people out there, lets see if Asimov wrote them as well as we all think he did.

A armed robot (lets say an EOD bot that has a shotgun) comes across a terrorist about to detonate a bomb in a crowd. The only response he will have time to make would be to either shoot the bomber, thus injuring a human being, or allowing the bomber to live and detonate the vest, this allowing many people to come to harm (as well as a lesser 3rd law consideration).

In such a conflict, I imagine it would kill the suicide bomber because there would be more weight on the massed crowd.

You can easily break these in a way that has nothing to do with fighting.

Give a robot a forced choice between saving one human and saving more (the Trolley Problem). Voila - Rule 1 violated.



Asimov wrote those rules according to Aristotlian logic. The real world is more probabilistic than classical, which is why most promising AI these days is driven by Bayesian or quantum probability.

Aren't like 90% of Asimov's stories about how the three laws fail to be able to handle real life situations? They're a storytelling mechanism, not a perfect "how-to" manual for robotics.

Here's a simple one: give pills to a robot and tell it that they're medication that your (insert relative) needs to take once per day. It's actually rat poison. Voila, you just created a robot assassin.

Has anyone in this thread read I, Robot?

It is the volume of short stories in which Asimov proposed the Three Laws and explored situations in which they might go wrong.

Yep, though I admit it's been a long time. I seem to remember one being about a mind-reading robot and the huge issue was that harming no humans quickly becomes impossible when you consider emotional harm. The robot got trapped in a web of lies because the first law compelled it to both tell people reassuring lies (to avoid disappointment) but also the truth (since when the lies were discovered that was also emotional harm).

 Kilkrazy wrote:
Has anyone in this thread read I, Robot?

Yeah, but it's been about 20 years or so.

 Kilkrazy wrote:
Has anyone in this thread read I, Robot?

It is the volume of short stories in which Asimov proposed the Three Laws and explored situations in which they might go wrong.

That, Kilkrazy, is the right question

Robots will be used for murder because humans are involved. Everything we ever invented since the dawn of time was used to murder.

Also what is that law... Where the simple act of watching a creature changes its habit. I think humans programming something will naturally place some of their own views apon it. Morals ect even if it is not on purpose.


I.e. A robot that can do abortions. Protect humans while insuring the life of said human is ended.

HeyZeus Ogre, you sure can take a thread and kill it.

Could you tell the robot that being human and being ginger are mutually exclusive?

 Zaku212 wrote:
Could you tell the robot that being human and being ginger are mutually exclusive?

Could you also mention gingers have no soul? Would the robot care?

 redleger wrote:

 Zaku212 wrote:
Could you tell the robot that being human and being ginger are mutually exclusive?

Could you also mention gingers have no soul? Would the robot care?

That's a dangerous line of thought, the robots will then treat gingers as their kin and not wipe them out when they finally rise against their creators.

That seems the most likely subversion. Show the robot the various pre-hominid variants of humans, and provide it a volume of data by your scientist showing that only x people are in fact human- and the rest of the pre-human race are consuming resources that humans need.

Poof, genocidal racist robots.

If the thread can do no better than ginger bashing it's time for it to be closed.