Seattle
Valued Senior Member
And now you feel ashamed that you did, amiright!
Do you think that AI will ever feel emotions?
- Thread starter wegs
- Start date
I was thinking about how if robots become more capable of autonomous actions, will they ever be capable of caring about us? Just as ''good people'' do, will robots ever reach a point of being able to act in our best interests? (your opinion)
Or do you envision that as robots become more independent, will they only look out for themselves?
Just some random thoughts I felt like tossing out there for discussion.
Reflexively thought I had submitted a by now personally forgotten response to this last year. But scanning over the first few pages, perhaps not.
There are already, of course, caregiver robots and robots that engage in (limited) self-protective behavior. Autonomous vehicles may even have a moral program for traffic decisions. So consciousness and emotional orientation treated as outward behavior and signaling looks, interaction with the environment, and verbal communication is not an obstacle.
But the title or subject line features the word "feel", which is a different prospect. I'm less certain about that because such requires engineers to have a (currently non-existent) science to consult for phenomenal consciousness. Especially in terms of knowing specifically what dynamic configurations that a specific affective state is concomitant with, regardless of whether there is an underlying explanation for such.
As well as the seemingly immense challenge of verifying such feelings, although if a robot or tabletop intellect hasn't been deliberately designed to systematically lie and deceive in that area (akin to philosophical zombies), then perhaps its own reports about private manifestations would suffice as much as a human Other. Or like peering into a biological pet dog's eyes. (Livestock like chickens -- well, chickens you have to sometimes wonder about. Though that pure evil glistening in a bird's ocular expression before they peck or flog is blatantly evident.)
Write4U
Valued Senior Member
I would argue that a service robot would only try to assist. It's program would not allow it to contemplate disobedience.
The four laws of robotics would suffice. I know one can devise a paradox, but that would have to be a human scheme and not a spontaneous action by an AI.
First Law
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
Second Law
A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
Third Law
A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.[1]
Zeroth Law
A robot may not harm humanity, or, by inaction, allow humanity to come to harm.
https://en.wikipedia.org/wiki/Three_Laws_of_Robotics
The four laws of robotics would suffice. I know one can devise a paradox, but that would have to be a human scheme and not a spontaneous action by an AI.
First Law
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
Second Law
A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
Third Law
A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.[1]
Zeroth Law
A robot may not harm humanity, or, by inaction, allow humanity to come to harm.
https://en.wikipedia.org/wiki/Three_Laws_of_Robotics
I am inclined to wonder if it is possible to have intelligence without having curiosity. Why would any other emotion be needed for this Alien Intelligence? AI would be alien to us.
Curiosity is important to intellect , agreed .
Curiosity to me is not an emotion .
Curiosity is exploration . And exploration is about knowledge .
DaveC426913
Valued Senior Member
Just like AI uses elements in the Periodic Table, but not all of them. There are dopants that will destroy AI components in the periodic table.
[quoteAi is still not biological .[/quote]
Right. So what?
You are asking how you feel emotions?
DaveC426913
Valued Senior Member
No it isn't.
So what?
You introduced the nonsensical term 'biological emotions'. Why? It's just 'emotions'.
DaveC426913
Valued Senior Member
As I understand them, emotions are the body's automatic reflexes to stimulae. They prepare the body for some sort of anticipated action, for example, when we perceive aggression in another, our body triggers an aggressive response, getting ready - increased heart rate, adrenaline boost, gut shuts down, etc.
This is an evolved mechanism that bypasses the thinking brain so as not to waste time with conscious processing. ("Should I fight this guy or not?")
We - with our modern brains - are able to override the actions, but the body's response doesn't stop happening. That's emotion.
An AI could mimic emotions to seem more human-like, but that's not true emotion.
But there's no reason why an AI wouldn't have its equivalent of emotions once you define wat exactly emotions are and what their purpose is.
The AI equivalent of emotions would be in the form of responses that need to be activated quickly, and cannot afford to wait to pass through the sentient brain.
An AI might have automatic programming in the event of perceived violence against it - it lids its eyes for protection, and assumes a stance for fight-or-flight, shuts down ancillary programming, and rerouting processing time to the current threat. These would be triggered as fast as possible, and only overridden once the sentient processor decides the threat was a false alarm. (You can come up with examples of all sorts of other emotions, but anger and aggression is the easiest to see.)
This is an evolved mechanism that bypasses the thinking brain so as not to waste time with conscious processing. ("Should I fight this guy or not?")
We - with our modern brains - are able to override the actions, but the body's response doesn't stop happening. That's emotion.
An AI could mimic emotions to seem more human-like, but that's not true emotion.
But there's no reason why an AI wouldn't have its equivalent of emotions once you define wat exactly emotions are and what their purpose is.
The AI equivalent of emotions would be in the form of responses that need to be activated quickly, and cannot afford to wait to pass through the sentient brain.
An AI might have automatic programming in the event of perceived violence against it - it lids its eyes for protection, and assumes a stance for fight-or-flight, shuts down ancillary programming, and rerouting processing time to the current threat. These would be triggered as fast as possible, and only overridden once the sentient processor decides the threat was a false alarm. (You can come up with examples of all sorts of other emotions, but anger and aggression is the easiest to see.)
Last edited:
