Ordering and stealing robots

Terry Austin wrote:

> Tim Bruening <tsbrueni.RemoveThis@pop.dcn.davis.ca.us> wrote in
> news:44ECD044.49D005CE@pop.dcn.davis.ca.us:
>
> >
> >
> > John Reiher wrote:
> >
> >> In article <44ECC1D9.973F446D.RemoveThis@pop.dcn.davis.ca.us>,
> >> Tim Bruening <tsbrueni.RemoveThis@pop.dcn.davis.ca.us> wrote:
> >>
> >> > John Reiher wrote:
> >> >
> >> > > In article <Xns9826A33BBAAC5taustingmail.RemoveThis@216.168.3.64>,
> >> > > No 33 Secretary <terry.notaniceperson.RemoveThis@gmail.com> wrote:
> >> > >
> >> > > > "norrin" <adweiland.RemoveThis@hotmail.com> wrote in
> >> > > > news:1156200516.996024.73280@m73g2000cwd.googlegroups.com:
> >> > > >
> >> > > > >
> >> > > > > Tim Bruening wrote:
> >> > > > >> Stan Pope wrote:
> >> > > > >>
> >> > > > >> > "Joseph Nebus" wrote;
> >> > > > >> >
> >> > > > >> > > It seems reasonable to me that a robot could have
> >> > > > >> > > impressed into it that there are dire consequences to
> >> > > > >> > > not being where its legal owner has ordered it to be
> >> > > > >> >
> >> > > > >> > I seem to recall something about a hierarchy of Second Law
> >> > > > >> > priorities. A rightful owner's orders, when not
> >> > > > >> > conflicting with a First Law requirement would override a
> >> > > > >> > non-owners orders. The reason being that, absent any
> >> > > > >> > other First Law issues, a robot would by default do harm
> >> > > > >> > to it's owner by not following their orders, but
> >> > > > >> > would not do harm to a non-owner by ignoring their orders.
> >> > > > >> >
> >> > > > >> > But an action required by human laws would override the
> >> > > > >> > orders of a robot's owner if those orders violate a law,
> >> > > > >> > not only because the top priority Second Law orders are
> >> > > > >> > human laws themselves, but because a human owner too is
> >> > > > >> > obligated to follow human laws. And allowing its owner to
> >> > > > >> > break a human law would be a violation of the
> >> > > > >> > robot's First Law since violating a human law is, by
> >> > > > >> > default, harmful to that human. From this it would follow
> >> > > > >> > that a robot should protect itself from theft on the
> >> > > > >> > grounds of Second Law requirements of its owner AND on the
> >> > > > >> > grounds of the importance of human laws which allow it to
> >> > > > >> > ignore orders of a potential thief since those orders
> >> > > > >> > would violate a greater human law (or order).
> >> > > > >> > In fact, a robot would also be following the First Law by
> >> > > > >> > protecting the potential thief from harm by not allowing
> >> > > > >> > that thief to break a law.
> >> > > > >> >
> >> > > > >> > So there are multiple methods of Robotic Law reasoning for
> >> > > > >> > a robot to not only give top priority to it's owner's
> >> > > > >> > orders, but to only follow "legal" orders by anyone, and
> >> > > > >> > for a robot to guard itself from theft.
> >> > > > >>
> >> > > > >> What if the thief says "Follow me or I'll kill myself!"?
> >> > > > >
> >> > > > > Subject is suicidal. Subject must be restrained to prevent
> >> > > > > self-harm. If subject is able to speak, subject will try to
> >> > > > > prevent this.
> >> > > > >
> >> > > > > BONK!
> >> > > > >
> >> > > > > Contacts security.
> >> > > > >
> >> > > > How about "follow me or someone who is not here right now will
> >> > > > murder my family, and I have proof this is true."
> >> > >
> >> > > "I am contacting the authorities and I am informing them of this
> >> > > situation. I have transmitted your likeness to them their
> >> > > computers and your identification and residence will be
> >> > > determined. Do not worry sir, your family will soon be safe..."
> >> >
> >> > "Follow and obey me, and do not contact the police, or someone who
> >> > isn't hear right now
> >> > (such as an Al-Qaida terrorist) will murder my family".
> >>
> >> As an emotional argument, that would work on a person. But a robot
> >> governed by the Three Laws isn't emotional, it is entirely logical.
> >> So the response he would give is same as I had originally written.
> >> The best course of action in this situation is to contact the
> >> authorities and place you and your family under protection.
> >>
> >> Why? Because the first law is going to overrule the second law every
> >> time. The robot must protect you and your family from harm, and even
> >> if you convince it not to notify the police, it will probably contact
> >> every robot within broadcast distance and have them find out where
> >> you family is and place them under protection, while it works out a
> >> way to place you under protection.
> >>
> >> You have to come up with a way to convince a robot to obey you
> >> without placing a human, real or imagined, into harm. Otherwise first
> >> law kicks in and overrides the second law.
> >
> > "I am on an urgent mission to stop an Al-Qaida plot to destroy the
> > Empire State Building, and I need your help!".
> >
> "There is a corrupt police captain holding my family hostage. If you
> contact the police, or any other authorities, my family will be murdered
> directly because of your actions."
>
> John's really not very good at this game. Seems to think the Three Laws
> are something to be taken seriously, without an objective definition of
> "harm" build in to them. Which takes us back to your robot slowly
> torturing you to death to martyr you for your faith, to ensure you
> entrance to heaven.

Cutie is the only type of Asimov robot I can imagine doing such a thing.
 >> Stay informed about: Ordering and stealing robots 

Tim Bruening <tsbrueni DeleteThis @pop.dcn.davis.ca.us> wrote in
news:4754DAA9.6FEB11EE@pop.dcn.davis.ca.us:

>
>
> Terry Austin wrote:
>
>> Tim Bruening <tsbrueni DeleteThis @pop.dcn.davis.ca.us> wrote in
>> news:44ECD044.49D005CE@pop.dcn.davis.ca.us:
>>
>> >
>> >
>> > John Reiher wrote:
>> >
>> >> In article <44ECC1D9.973F446D DeleteThis @pop.dcn.davis.ca.us>,
>> >> Tim Bruening <tsbrueni DeleteThis @pop.dcn.davis.ca.us> wrote:
>> >>
>> >> > John Reiher wrote:
>> >> >
>> >> > > In article <Xns9826A33BBAAC5taustingmail DeleteThis @216.168.3.64>,
>> >> > > No 33 Secretary <terry.notaniceperson DeleteThis @gmail.com> wrote:
>> >> > >
>> >> > > > "norrin" <adweiland DeleteThis @hotmail.com> wrote in
>> >> > > > news:1156200516.996024.73280@m73g2000cwd.googlegroups.com:
>> >> > > >
>> >> > > > >
>> >> > > > > Tim Bruening wrote:
>> >> > > > >> Stan Pope wrote:
>> >> > > > >>
>> >> > > > >> > "Joseph Nebus" wrote;
>> >> > > > >> >
>> >> > > > >> > > It seems reasonable to me that a robot could have
>> >> > > > >> > > impressed into it that there are dire consequences to
>> >> > > > >> > > not being where its legal owner has ordered it to be
>> >> > > > >> >
>> >> > > > >> > I seem to recall something about a hierarchy of Second
>> >> > > > >> > Law priorities. A rightful owner's orders, when not
>> >> > > > >> > conflicting with a First Law requirement would override
>> >> > > > >> > a non-owners orders. The reason being that, absent any
>> >> > > > >> > other First Law issues, a robot would by default do
>> >> > > > >> > harm to it's owner by not following their orders, but
>> >> > > > >> > would not do harm to a non-owner by ignoring their
>> >> > > > >> > orders.
>> >> > > > >> >
>> >> > > > >> > But an action required by human laws would override the
>> >> > > > >> > orders of a robot's owner if those orders violate a
>> >> > > > >> > law, not only because the top priority Second Law
>> >> > > > >> > orders are human laws themselves, but because a human
>> >> > > > >> > owner too is obligated to follow human laws. And
>> >> > > > >> > allowing its owner to break a human law would be a
>> >> > > > >> > violation of the robot's First Law since violating a
>> >> > > > >> > human law is, by default, harmful to that human. From
>> >> > > > >> > this it would follow that a robot should protect itself
>> >> > > > >> > from theft on the grounds of Second Law requirements of
>> >> > > > >> > its owner AND on the grounds of the importance of human
>> >> > > > >> > laws which allow it to ignore orders of a potential
>> >> > > > >> > thief since those orders would violate a greater human
>> >> > > > >> > law (or order). In fact, a robot would also be
>> >> > > > >> > following the First Law by protecting the potential
>> >> > > > >> > thief from harm by not allowing that thief to break a
>> >> > > > >> > law.
>> >> > > > >> >
>> >> > > > >> > So there are multiple methods of Robotic Law reasoning
>> >> > > > >> > for a robot to not only give top priority to it's
>> >> > > > >> > owner's orders, but to only follow "legal" orders by
>> >> > > > >> > anyone, and for a robot to guard itself from theft.
>> >> > > > >>
>> >> > > > >> What if the thief says "Follow me or I'll kill myself!"?
>> >> > > > >
>> >> > > > > Subject is suicidal. Subject must be restrained to
>> >> > > > > prevent self-harm. If subject is able to speak, subject
>> >> > > > > will try to prevent this.
>> >> > > > >
>> >> > > > > BONK!
>> >> > > > >
>> >> > > > > Contacts security.
>> >> > > > >
>> >> > > > How about "follow me or someone who is not here right now
>> >> > > > will murder my family, and I have proof this is true."
>> >> > >
>> >> > > "I am contacting the authorities and I am informing them of
>> >> > > this situation. I have transmitted your likeness to them their
>> >> > > computers and your identification and residence will be
>> >> > > determined. Do not worry sir, your family will soon be
>> >> > > safe..."
>> >> >
>> >> > "Follow and obey me, and do not contact the police, or someone
>> >> > who isn't hear right now
>> >> > (such as an Al-Qaida terrorist) will murder my family".
>> >>
>> >> As an emotional argument, that would work on a person. But a robot
>> >> governed by the Three Laws isn't emotional, it is entirely
>> >> logical. So the response he would give is same as I had originally
>> >> written. The best course of action in this situation is to contact
>> >> the authorities and place you and your family under protection.
>> >>
>> >> Why? Because the first law is going to overrule the second law
>> >> every time. The robot must protect you and your family from harm,
>> >> and even if you convince it not to notify the police, it will
>> >> probably contact every robot within broadcast distance and have
>> >> them find out where you family is and place them under protection,
>> >> while it works out a way to place you under protection.
>> >>
>> >> You have to come up with a way to convince a robot to obey you
>> >> without placing a human, real or imagined, into harm. Otherwise
>> >> first law kicks in and overrides the second law.
>> >
>> > "I am on an urgent mission to stop an Al-Qaida plot to destroy the
>> > Empire State Building, and I need your help!".
>> >
>> "There is a corrupt police captain holding my family hostage. If you
>> contact the police, or any other authorities, my family will be
>> murdered directly because of your actions."
>>
>> John's really not very good at this game. Seems to think the Three
>> Laws are something to be taken seriously, without an objective
>> definition of "harm" build in to them. Which takes us back to your
>> robot slowly torturing you to death to martyr you for your faith, to
>> ensure you entrance to heaven.
>
> I am now imagining robots Crucifying Jesus to redeem the whole human
> race.
>
You've heard of Nancy Luft, then, and her time traveling sputniks.

--
Terry Austin
Beware the other head of science. It bites.
 >> Stay informed about: Ordering and stealing robots 

In article <Xns99FCE069B9447taustingmail DeleteThis @216.168.3.64>,
Terry Austin <terry.notaniceperson DeleteThis @gmail.com> wrote:

> Tim Bruening <tsbrueni DeleteThis @pop.dcn.davis.ca.us> wrote in
> news:4754DAA9.6FEB11EE@pop.dcn.davis.ca.us:
>
> >
> >
> > Terry Austin wrote:
> >
> >> Tim Bruening <tsbrueni DeleteThis @pop.dcn.davis.ca.us> wrote in
> >> news:44ECD044.49D005CE@pop.dcn.davis.ca.us:
> >>
> >> >
> >> >
> >> > John Reiher wrote:
> >> >
> >> >> In article <44ECC1D9.973F446D DeleteThis @pop.dcn.davis.ca.us>,
> >> >> Tim Bruening <tsbrueni DeleteThis @pop.dcn.davis.ca.us> wrote:
> >> >>
> >> >> > John Reiher wrote:
> >> >> >
> >> >> > > In article <Xns9826A33BBAAC5taustingmail DeleteThis @216.168.3.64>,
> >> >> > > No 33 Secretary <terry.notaniceperson DeleteThis @gmail.com> wrote:
> >> >> > >
> >> >> > > > "norrin" <adweiland DeleteThis @hotmail.com> wrote in
> >> >> > > > news:1156200516.996024.73280@m73g2000cwd.googlegroups.com:
> >> >> > > >
> >> >> > > > >
> >> >> > > > > Tim Bruening wrote:
> >> >> > > > >> Stan Pope wrote:
> >> >> > > > >>
> >> >> > > > >> > "Joseph Nebus" wrote;
> >> >> > > > >> >
> >> >> > > > >> > > It seems reasonable to me that a robot could have
> >> >> > > > >> > > impressed into it that there are dire consequences to
> >> >> > > > >> > > not being where its legal owner has ordered it to be
> >> >> > > > >> >
> >> >> > > > >> > I seem to recall something about a hierarchy of Second
> >> >> > > > >> > Law priorities. A rightful owner's orders, when not
> >> >> > > > >> > conflicting with a First Law requirement would override
> >> >> > > > >> > a non-owners orders. The reason being that, absent any
> >> >> > > > >> > other First Law issues, a robot would by default do
> >> >> > > > >> > harm to it's owner by not following their orders, but
> >> >> > > > >> > would not do harm to a non-owner by ignoring their
> >> >> > > > >> > orders.
> >> >> > > > >> >
> >> >> > > > >> > But an action required by human laws would override the
> >> >> > > > >> > orders of a robot's owner if those orders violate a
> >> >> > > > >> > law, not only because the top priority Second Law
> >> >> > > > >> > orders are human laws themselves, but because a human
> >> >> > > > >> > owner too is obligated to follow human laws. And
> >> >> > > > >> > allowing its owner to break a human law would be a
> >> >> > > > >> > violation of the robot's First Law since violating a
> >> >> > > > >> > human law is, by default, harmful to that human. From
> >> >> > > > >> > this it would follow that a robot should protect itself
> >> >> > > > >> > from theft on the grounds of Second Law requirements of
> >> >> > > > >> > its owner AND on the grounds of the importance of human
> >> >> > > > >> > laws which allow it to ignore orders of a potential
> >> >> > > > >> > thief since those orders would violate a greater human
> >> >> > > > >> > law (or order). In fact, a robot would also be
> >> >> > > > >> > following the First Law by protecting the potential
> >> >> > > > >> > thief from harm by not allowing that thief to break a
> >> >> > > > >> > law.
> >> >> > > > >> >
> >> >> > > > >> > So there are multiple methods of Robotic Law reasoning
> >> >> > > > >> > for a robot to not only give top priority to it's
> >> >> > > > >> > owner's orders, but to only follow "legal" orders by
> >> >> > > > >> > anyone, and for a robot to guard itself from theft.
> >> >> > > > >>
> >> >> > > > >> What if the thief says "Follow me or I'll kill myself!"?
> >> >> > > > >
> >> >> > > > > Subject is suicidal. Subject must be restrained to
> >> >> > > > > prevent self-harm. If subject is able to speak, subject
> >> >> > > > > will try to prevent this.
> >> >> > > > >
> >> >> > > > > BONK!
> >> >> > > > >
> >> >> > > > > Contacts security.
> >> >> > > > >
> >> >> > > > How about "follow me or someone who is not here right now
> >> >> > > > will murder my family, and I have proof this is true."
> >> >> > >
> >> >> > > "I am contacting the authorities and I am informing them of
> >> >> > > this situation. I have transmitted your likeness to them their
> >> >> > > computers and your identification and residence will be
> >> >> > > determined. Do not worry sir, your family will soon be
> >> >> > > safe..."
> >> >> >
> >> >> > "Follow and obey me, and do not contact the police, or someone
> >> >> > who isn't hear right now
> >> >> > (such as an Al-Qaida terrorist) will murder my family".
> >> >>
> >> >> As an emotional argument, that would work on a person. But a robot
> >> >> governed by the Three Laws isn't emotional, it is entirely
> >> >> logical. So the response he would give is same as I had originally
> >> >> written. The best course of action in this situation is to contact
> >> >> the authorities and place you and your family under protection.
> >> >>
> >> >> Why? Because the first law is going to overrule the second law
> >> >> every time. The robot must protect you and your family from harm,
> >> >> and even if you convince it not to notify the police, it will
> >> >> probably contact every robot within broadcast distance and have
> >> >> them find out where you family is and place them under protection,
> >> >> while it works out a way to place you under protection.
> >> >>
> >> >> You have to come up with a way to convince a robot to obey you
> >> >> without placing a human, real or imagined, into harm. Otherwise
> >> >> first law kicks in and overrides the second law.
> >> >
> >> > "I am on an urgent mission to stop an Al-Qaida plot to destroy the
> >> > Empire State Building, and I need your help!".
> >> >
> >> "There is a corrupt police captain holding my family hostage. If you
> >> contact the police, or any other authorities, my family will be
> >> murdered directly because of your actions."
> >>
> >> John's really not very good at this game. Seems to think the Three
> >> Laws are something to be taken seriously, without an objective
> >> definition of "harm" build in to them. Which takes us back to your
> >> robot slowly torturing you to death to martyr you for your faith, to
> >> ensure you entrance to heaven.
> >
> > I am now imagining robots Crucifying Jesus to redeem the whole human
> > race.
> >
> You've heard of Nancy Luft, then, and her time traveling sputniks.

I stopped playing the game when I realized that we weren't dealing with
a robot, but with a person playing at a robot. You want to steal a
robot? Real simple: Walk up to it and tell it "Robot, turn yourself off."

The robot replies, "Why should I turn myself off?"

You reply, "You do not need to know why. Turn yourself off, now."

There may be some back and forth, but eventually the robot will shut
itself down. Then, if you're familiar with the model, you just walk up
to it and engage the manual shutdown so it can't turn on by itself. Get
a dolly and you have a robot.

No need for complex logical arguments, you just tell it to shut down, no
other reasons are necessary. Similarly, all you need to do is tell it to
follow you and refuse to tell it why other than it doesn't need to know.
At that point you're reinforcing the Second Law without invoking the
First Law at all.

--
The Kedamono Dragon
Pull Pinky's favorite words to email me.
http://www.ahtg.net
Have Mac, will Compute

Check out the PowerPointers Shop at:
http://www.cafeshops.com/PowerPointers

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----
 >> Stay informed about: Ordering and stealing robots