Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
Was reading an article about an engineer who was fired by google because he claimed their voice technology has developed sentience ( the ability to feel ).
Which got me to thinking.
Would robots with problem solving ability and sentience " evolve " from scratch differently, depending upon their primary objective.
For example, one objective could be it's own survival, another be human survival.
Or more intersting to me, what if the robot had no primary objective, but just a blank slate with sentience, and the ability to learn.
Would it develope compassion and empathy, on it's own?
Was reading an article about an engineer who was fired by google because he claimed their voice technology has developed sentience ( the ability to feel ).
Which got me to thinking.
Would robots with problem solving ability and sentience " evolve " from scratch differently, depending upon their primary objective.
For example, one objective could be it's own survival, another be human survival.
Or more intersting to me, what if the robot had no primary objective, but just a blank slate with sentience, and the ability to learn.
Would it develope compassion and empathy, on it's own?
Hoping someone smarter than me will chime in here
I can't answer your question at all, but I'll toss into the conversation about sentience, that this is a topic often discussed in Buddhist circle. And the reason it is, is because Buddhism talks about not taking the life of a sentient being. And the big debate point is how sentience relates to animals that become dinner. (And this is why virtually all butchers in Thailand are not Buddhists, but are instead Muslims...although Buddhists seek to cheat on the concept and eat the food since they didn't kill the food...including the monks).
I can't answer your question at all, but I'll toss into the conversation about sentience, that this is a topic often discussed in Buddhist circle. And the reason it is, is because Buddhism talks about not taking the life of a sentient being. And the big debate point is how sentience relates to animals that become dinner. (And this is why virtually all butchers in Thailand are not Buddhists, but are instead Muslims...although Buddhists seek to cheat on the concept and eat the food since they didn't kill the food...including the monks).
I believe animals should be included in that as well which causes a problem because I buy mass produced animal products. It's starting to really bother me for some reason.
Was reading an article about an engineer who was fired by google because he claimed their voice technology has developed sentience ( the ability to feel ).
Which got me to thinking.
Would robots with problem solving ability and sentience " evolve " from scratch differently, depending upon their primary objective.
For example, one objective could be it's own survival, another be human survival.
The technology you are looking for here is neural networks, which are simple computer models of brains. These learn from data being repeatedly input, with expected results modifying the neural network. So if you programmed a neural network only for it's own survival, this could be bad for people unless you 1) also use data that also programs for human survival, 2) use data that shows robot survival depends on human survival, or 3), have a neural network that is so far advanced (if possible), it understands humans are also sentient.
My worry is that such a network may learn things from data that we do not know is there, or allow it to find it's own source to learn from. Imagine a robot that learns what humans are doing to the planet. Having learnt from data, it does not have any specific rules about moral decisions other than what it has learnt from that data.
Fortunately the neural networks I train are not that complex, and are used for specific jobs such as game controllers or predicting what populations will usually buy in supermarkets based on what they have previously bought. For example, people who buy diapers are also more likely to buy beer.
Quote:
Originally Posted by Oakback
Or more intersting to me, what if the robot had no primary objective, but just a blank slate with sentience, and the ability to learn.
Would it develope compassion and empathy, on it's own?
Hoping someone smarter than me will chime in here
That would need to be a big neural network that develops an internal model of the real world, and develops the ability to experience feelings (qualia). As we do not know how animals produce Qualia, we do not know if robots can (although looking at the evidence, they probably would).
The question is, compassion and empathy for whom? Humans or other robots.
The technology you are looking for here is neural networks, which are simple computer models of brains. These learn from data being repeatedly input, with expected results modifying the neural network. So if you programmed a neural network only for it's own survival, this could be bad for people unless you 1) also use data that also programs for human survival, 2) use data that shows robot survival depends on human survival, or 3), have a neural network that is so far advanced (if possible), it understands humans are also sentient.
My worry is that such a network may learn things from data that we do not know is there, or allow it to find it's own source to learn from. Imagine a robot that learns what humans are doing to the planet. Having learnt from data, it does not have any specific rules about moral decisions other than what it has learnt from that data.
Fortunately the neural networks I train are not that complex, and are used for specific jobs such as game controllers or predicting what populations will usually buy in supermarkets based on what they have previously bought. For example, people who buy diapers are also more likely to buy beer.
That would need to be a big neural network that develops an internal model of the real world, and develops the ability to experience feelings (qualia). As we do not know how animals produce Qualia, we do not know if robots can (although looking at the evidence, they probably would).
The question is, compassion and empathy for whom? Humans or other robots.
Unlike you I do not do that kind of software development but know enough about it to be interested / dangerous. It is amazing though what I can accomplish just by generalizing rule systems, to do things like correct and standardize business names and addresses in ways that can be controlled by non-programmers.
To your points and the OP's questions, though, I think that much, if not everything, hinges on definitions of things like "sentience", "feelings", "qualia", etc. IMO the fired employee's definitions set too low of a bar. But it does raise interesting questions, similar to the "Turing test": if most observers can't tell simulated awareness from actual awareness, does it really matter if the AI has actual feelings or is experiencing actual qualia, as to the question of whether or not it is sentient? We already know that people tend to anthropomorphize AIs, or if they are embodied in a dog-like robot, to ... what would the term be ... "canineize" them. All these systems have to do is put on a convincing enough, accurate enough show and we'll accept them as such. And maybe if we start doing that it is harmful to us to get used to the notion of treating them badly or even indifferently, as it will cause us to start doing the same to other humans.
I believe animals should be included in that as well which causes a problem because I buy mass produced animal products. It's starting to really bother me for some reason.
It bothered me for a while too...as a Buddhist...but then I finally decided to remember that virtually all Thai Buddhist monks eat meat if it is given to them. And I don't quite buy into their 'well, we didn't kill the animal ourselves, so it's not an issue for our karma' thinking. What I have done is reduced my meat intake to one meal a day almost every day. That's best I can do.
The technology you are looking for here is neural networks, which are simple computer models of brains. These learn from data being repeatedly input, with expected results modifying the neural network. So if you programmed a neural network only for it's own survival, this could be bad for people unless you 1) also use data that also programs for human survival, 2) use data that shows robot survival depends on human survival, or 3), have a neural network that is so far advanced (if possible), it understands humans are also sentient.
My worry is that such a network may learn things from data that we do not know is there, or allow it to find it's own source to learn from. Imagine a robot that learns what humans are doing to the planet. Having learnt from data, it does not have any specific rules about moral decisions other than what it has learnt from that data.
Fortunately the neural networks I train are not that complex, and are used for specific jobs such as game controllers or predicting what populations will usually buy in supermarkets based on what they have previously bought. For example, people who buy diapers are also more likely to buy beer.
That would need to be a big neural network that develops an internal model of the real world, and develops the ability to experience feelings (qualia). As we do not know how animals produce Qualia, we do not know if robots can (although looking at the evidence, they probably would).
The question is, compassion and empathy for whom? Humans or other robots.
Learned a new word; QUALIA. Never heard it.
Is it different then " sentience" ? Qualia seems more like just a response to stimuli. Is that right.
I'm wondering would compassion or empathy be any different if it was for a robot or human. Or both at the same time.
Fascinating topic for me as a theist. Raises questions regarding the act of creation or creating. Questions of consciousness. What it is to be fully human.
It bothered me for a while too...as a Buddhist...but then I finally decided to remember that virtually all Thai Buddhist monks eat meat if it is given to them. And I don't quite buy into their 'well, we didn't kill the animal ourselves, so it's not an issue for our karma' thinking. What I have done is reduced my meat intake to one meal a day almost every day. That's best I can do.
Taking steps toward it seems like a good idea. Thanks
It bothered me for a while too...as a Buddhist...but then I finally decided to remember that virtually all Thai Buddhist monks eat meat if it is given to them. And I don't quite buy into their 'well, we didn't kill the animal ourselves, so it's not an issue for our karma' thinking. What I have done is reduced my meat intake to one meal a day almost every day. That's best I can do.
I spent some time as a vegetarian for multiple reasons. Health, enviornmental, and I felt hypocritical as an animal lover.
Try to limit red meat intake now to once or twice a week, but it still feels like half measures.
May go back to being a vegetarian.
But a good prime rib eye is ....well....yummy.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
