Appreciate if someone could help me open the video.
------------

Thank you cheni

This message has been edited. Last edited by: Rustyblade,

Can we have a choice of No without Asimov's bleating attached?

quote:

Originally posted by Sig2340:
Can we have a choice of No without Asimov's bleating attached?

Good point. Thank you.

With current state of human intelligence and competencies, NO to AI doing anything that might harm humans (intentionally or not) without oversight.

I don't have a problem with autonomous vehicles attacking and destroying military vehicles during combat, with combatants within or not. But any device targeting individuals must have direct oversight.

On another note, I am shocked that there has not been some kind of terror attack utilizing a drone. It would be a simple matter to load a large, weight bearing drone up with a toxic substance, biological weapon, or radioactive threat. It would be simple to pack one with explosives and fly it into the flight path of a jet on approach. The variety of attack scenarios are infinite.

"Once the rock has left your hand, the only oversight you have is as a spectator."

-Ook the caveman

I think what we might say has no bearing on what will actually be done.

quote:

...But AI must obey Asimov's 3 rules of robotics

In case you have never read them:

"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."
.

Are you kidding me? Just the fact that you linked to the right arm of the beast freaks me out!

Asimov was brilliant, as were his Robot books, but although his three laws impressed the ever-lovin’ out of me as a kid, others pointed out later that even if it were possible to ensure they were enforced faithfully in AI creations, that wouldn’t guarantee things would turn out the way we humans would prefer. One of the most obvious issues is the “or, through inaction, allow a human to come to harm” clause. An AI might decide that countless things we humans do cause harm to ourselves: owning weapons, eating the wrong foods, releasing pollutants into the environment, engaging in risky activities such as riding motorcycles, etc. And because controlling such things would require rigorous control over our behaviors, that’s what would be necessary: countless free will restrictions.

If the three laws were followed, it would of course be impossible for an AI to serve as a weapon except, possibly, against other robots.

I welcome our Robot overlords.

I'm getting sick of people and their "feelings" being so fucking important.

I'm sure South Korea wouldn't mind the Samsung automated gun platform system gunning anything in sight down if it's northern neighbors were to march across the dmz.

Let them loose on our foes. I have worked in robotics for 30 years and with the proper rules then no worries. Don’t fall for the Hollywood BS terminator crap.

quote:

No -- AI has no restraints

Not what I had in mind.

It should be:

No -- AI cannot be used as a weapon against any living entity, ever. Period. No exceptions.

quote:

Originally posted by arcwelder76:
I welcome our Robot overlords.

I'm getting sick of people and their "feelings" being so fucking important.

Agreed. I'd rather depend on AI than people in most cases.

I think this made the rounds about 6 months ago.

quote:

Originally posted by BBMW:
I think what we might say has no bearing on what will actually be done.

Agree with this. Regardless of what we as a people with a conscious might do, our enemies will use it to fight us without the moral constraints.

How is an AI killing someone without human supervision any different than a non-AI land mine or booby trap killing someone? Dead is dead.