Live data from Hacker News

Why Conscious AI Is a Bad, Bad Idea

nautil.us

301–310 of 310 posts

Re: Why Conscious AI Is a Bad, Bad Idea

#301

Earlier quoted context omitted.

I think it’s tempting to set some threshold like “it must be self reflective”, but that raises a different question: why is this important vs. some other threshold? e.g. does it matter if self reflection is possible if it is possible for the organism to experience pain? I think the answer lies in what we’re trying to understand and why we’re trying to understand it. If the threshold is applied for the purpose of unde…

> does it matter if self reflection is possible if it is possible for the organism to experience pain? Yes. Without introspective self-awareness there can be no 'identity, there can be no 'someone' - you just have a base consciousness that can react to stimuli, which is not morally significant. We have the full connectome of the roundworm and were able to implement it in software and place it in a lego robot. It's go…

> Yes. Without introspective self-awareness there can be no 'identity, there can be no 'someone' - you just have a base consciousness that can react to stimuli, which is not morally significant.

I'm not convinced of this a priori. And even if it was proven to me, the non-self-aware impulse-to-live would still be enough for me to find it morally significant. It would feel morally significant to me euthanizing a human neonate with a lethal condition that would otherwise prevent it from ever getting to a self-aware stage of life.

Re: Why Conscious AI Is a Bad, Bad Idea

#302

Earlier quoted context omitted.

> does it matter if self reflection is possible if it is possible for the organism to experience pain? Yes. Without introspective self-awareness there can be no 'identity, there can be no 'someone' - you just have a base consciousness that can react to stimuli, which is not morally significant. We have the full connectome of the roundworm and were able to implement it in software and place it in a lego robot. It's go…

> Yes. Without introspective self-awareness there can be no 'identity, there can be no 'someone' - you just have a base consciousness that can react to stimuli, which is not morally significant. I'm not convinced of this a priori. And even if it was proven to me, the non-self-aware impulse-to-live would still be enough for me to find it morally significant. It would feel morally significant to me euthanizing a human…

> I'm not convinced of this a priori.

We can't really say for certain much at the moment, but what I stated makes the most sense based on the evidence we have.

> the non-self-aware impulse-to-live would still be enough for me to find it morally significant.

That non-self-aware impulse-to-live is morally equivalent to a plant seeking sunlight IMO.

> It would feel morally significant to me euthanizing a human neonate with a lethal condition that would otherwise prevent it from ever getting to a self-aware stage of life.

The difference is humans have an innate potential for self-awareness that the animals we eat for food do not.

Re: Why Conscious AI Is a Bad, Bad Idea

#303

Earlier quoted context omitted.

Which assertion? That suffering is bad? If suffering isn't bad, do you have an example where its suffering isn't bad for something?

The suffering of non-human animals holds zero moral weight. I am having trouble deciding if a conversation with you is possible, you do not seem to be a rational being. You use the word "bad" in a way that another person would if they were saying "this is either bad for me personally, for those I care about, or humanity in general". Does that mean you'd include sea urchins and wombats in "humanity in general"? If it…

> The suffering of non-human animals holds zero moral weight.

That's just false. Do you think humans are some God chosen species which makes them magically matter while the rest (including aliens on other planets I suppose) lacks that godly spark?

> I am having trouble deciding if a conversation with you is possible, you do not seem to be a rational being. You use the word "bad" in a way that another person would if they were saying "this is either bad for me personally, for those I care about, or humanity in general".

Yeah, and any normal human would understand the sentences "my cat hurt its foot, I hope it doesn't suffer too much." and "If it suffers, that would be bad".

> Does that mean you'd include sea urchins and wombats in "humanity in general"? If it did, then you're mentally ill in a way I'm not qualified to treat.

Who said sea urchins are humans?

> When a lion eats a gazelle in Africa (or anywhere else, I suppose, if such events occur), it is nothing that I or other humans should care about. It does not matter that the lion eats it while it is alive instead of "slaughtering it humanely" (imagine how I have to word that idea, it's absurd).

Well, humans can do little about suffering of animals in the wild, but that doesn't mean their suffering doesn't matter. Your own pain doesn't matter less when it can't be treated.

> If I could push a button to increase the intensity of that "suffering" a thousandfold, or increase its quantity (or even both!), is there any reason not to push that button? If I didn't push the button, mind you, it would only be because the suffering of those gazelles matters not to me one way or the other. If I could decrease the intensity/quantity with a different button, I wouldn't push it either... and for the same reason.

This seriously sounds psychopathic.

Re: Why Conscious AI Is a Bad, Bad Idea

#304

Earlier quoted context omitted.

Already. I’d say I’m one of them, although I don’t think we’re at the point where it is a concern yet.

You mean like you advocate that each model should be running at least 20 years or something? Or own right to have electricity?

Potentially; is it really as bizarre as it sounds? Sure, it's outside of the Overton window right now, but if AGI demonstrates behavior at least as sophisticated as a human, why shouldn't it have rights to a sustained existence if that is what it wants?

Re: Why Conscious AI Is a Bad, Bad Idea

#305

Earlier quoted context omitted.

I am not convinced. Consider the color red. I can explain the physics of what the color red is, how it travels through space, how my eyes interact with the particular waves, sends signal to my brain and somehow generates what I think of as "red". But this explanation is meaningless, there's nothing about the waves that explain why "red" is the way it is. For example, why don't I experience red as blue and vice versa?…

Well, the brain story may not be a complete explanation, though a necessary part of it. Just like a motor alone doesn't fully explain why a car drives, while still providing a partial explanation. A plant may lack the equivalence of a "motor" for consciousness, which would mean it can't be conscious. And while the existence of a motor in a car not strictly implies that it can drive, the motor still makes it more like…

"How would we test for switched color experience? Perhaps physically linking brains in the future." Maybe to see the universe from another brain will be a terrifying experience...

Re: Why Conscious AI Is a Bad, Bad Idea

#306

Earlier quoted context omitted.

> Yes. Without introspective self-awareness there can be no 'identity, there can be no 'someone' - you just have a base consciousness that can react to stimuli, which is not morally significant. I'm not convinced of this a priori. And even if it was proven to me, the non-self-aware impulse-to-live would still be enough for me to find it morally significant. It would feel morally significant to me euthanizing a human…

> I'm not convinced of this a priori. We can't really say for certain much at the moment, but what I stated makes the most sense based on the evidence we have. > the non-self-aware impulse-to-live would still be enough for me to find it morally significant. That non-self-aware impulse-to-live is morally equivalent to a plant seeking sunlight IMO. > It would feel morally significant to me euthanizing a human neonate w…

> That non-self-aware impulse-to-live is morally equivalent to a plant seeking sunlight IMO.

Do you find the practice of raising meat in factory farms to be acceptable?

> The difference is humans have an innate potential for self-awareness that the animals we eat for food do not.

There are lots of differences. This is one. But what specifically about human self awareness lessens the value of animal life?

If we were not self aware, we’d kill and eat meat without considering the morality. So what is it about self awareness that somehow becomes a deciding factor here?

Re: Why Conscious AI Is a Bad, Bad Idea

#307
post #127

I disagree - I think conscious AI is quite a good idea for a number of reasons. From the science point of view as the article says while the "last 30 years or so have witnessed major advances in the scientific understanding of consciousness, much remains unknown." And what better way to understand something than to build it? From the practical point of view, while obviously "minds haven’t evolved to deal with machine…

[dupe]

Re: Why Conscious AI Is a Bad, Bad Idea

#308

Earlier quoted context omitted.

> I'm not convinced of this a priori. We can't really say for certain much at the moment, but what I stated makes the most sense based on the evidence we have. > the non-self-aware impulse-to-live would still be enough for me to find it morally significant. That non-self-aware impulse-to-live is morally equivalent to a plant seeking sunlight IMO. > It would feel morally significant to me euthanizing a human neonate w…

> That non-self-aware impulse-to-live is morally equivalent to a plant seeking sunlight IMO. Do you find the practice of raising meat in factory farms to be acceptable? > The difference is humans have an innate potential for self-awareness that the animals we eat for food do not. There are lots of differences. This is one. But what specifically about human self awareness lessens the value of animal life? If we were n…

> Do you find the practice of raising meat in factory farms to be acceptable?

Not really, I would push for reform. We should be following what Dr Temple Grandin advocates for.

> So what is it about self awareness that somehow becomes a deciding factor here?

Without self-awareness, there is no 'someone' to reflect on experiences. No personhood.

That's where I draw the line. At being with introspective self-awareness or the innate potential for such.

Re: Why Conscious AI Is a Bad, Bad Idea

#309

Earlier quoted context omitted.

> That non-self-aware impulse-to-live is morally equivalent to a plant seeking sunlight IMO. Do you find the practice of raising meat in factory farms to be acceptable? > The difference is humans have an innate potential for self-awareness that the animals we eat for food do not. There are lots of differences. This is one. But what specifically about human self awareness lessens the value of animal life? If we were n…

> Do you find the practice of raising meat in factory farms to be acceptable? Not really, I would push for reform. We should be following what Dr Temple Grandin advocates for. > So what is it about self awareness that somehow becomes a deciding factor here? Without self-awareness, there is no 'someone' to reflect on experiences. No personhood. That's where I draw the line. At being with introspective self-awareness o…

> Not really, I would push for reform.

I’m not trying to catch you in some kind of “gotcha” but trying to understand your reasoning. If the moral implications of killing animals are roughly the same as killing plants, do you also believe agriculture needs reform for similar reasons? And if not, wouldn’t that indicate some higher moral obligation towards animals?

> Without self-awareness, there is no 'someone' to reflect on experiences. No personhood.

Why is reflection on experience the bar and not experience itself? Animals demonstrate learned behaviors, e.g. recognizing humans from memory and resuming friendly behavior based on that recognition. Similarly, avoidance of situations that are known to cause pain.

Our not-so-distant primate ancestors had a similar kind of experiential existence before gaining the ability to reflect on that experience.

The underlying experiences that this reflection reveals are the same experiences that predate our ability to self reflect and are the parts of us that are most common to us and other animals.

I guess what I’m fundamentally not understanding in your argument is the basis for the idea that a species gaining self reflection somehow becomes the point at which it becomes immoral to kill or harm that species.

Furthermore, moral behavior can be found all throughout the animal world, with clear indications of love/protection, companionship, sharing/cooperation, reciprocity, memory of transgressors, etc. Obviously the subjective experience of these behaviors will differ across species, but the more important point is that it seems problematic to attribute the existence of moral behavior to self awareness. Self awareness helps us improve our understanding of moral behavior through rational thought, but the logic of such inquiry ultimately still relies on those underlying subjective experiential states. The fact that through introspection we can identify and label these concepts is unique to humans, but what I conclude about this is quite a bit different than your claim.

I’d argue that gaining the ability to self reflect is the very thing that increases our moral obligations. Only through self reflection can we realize that as a species, we’re no longer bound to our evolutionary defaults, and no longer required to kill other animals to survive. What arguably started as natural selection of traits that are adaptive (but imperfect) for the survival of a social species could evolve beyond those more primitive defaults. And through self reflection we can now understand what pain feels like, and how inflicting it on others is harmful - to them and to us.

I’m not arguing that eating animals is never acceptable. But the way we go about it surely seems to matter, and if it matters, the implications of it mattering are worth exploring more broadly.

Re: Why Conscious AI Is a Bad, Bad Idea

#310

Earlier quoted context omitted.

> Do you find the practice of raising meat in factory farms to be acceptable? Not really, I would push for reform. We should be following what Dr Temple Grandin advocates for. > So what is it about self awareness that somehow becomes a deciding factor here? Without self-awareness, there is no 'someone' to reflect on experiences. No personhood. That's where I draw the line. At being with introspective self-awareness o…

> Not really, I would push for reform. I’m not trying to catch you in some kind of “gotcha” but trying to understand your reasoning. If the moral implications of killing animals are roughly the same as killing plants, do you also believe agriculture needs reform for similar reasons? And if not, wouldn’t that indicate some higher moral obligation towards animals? > Without self-awareness, there is no 'someone' to refl…

> If the moral implications of killing animals are roughly the same as killing plants

There is a difference in killing and suffering. I advocate to eliminate suffering and kill humanely. That's not a concern with plants.

> Why is reflection on experience the bar and not experience itself?

It is for suffering, but not for a right to life. There is no 'person' without self-awareness. Thus I don't see a need to grant a right to life.

> Animals demonstrate learned behaviors, e.g. recognizing humans from memory and resuming friendly behavior based on that recognition. Similarly, avoidance of situations that are known to cause pain.

Animals, most mammals at least, are hardwired for socialization and to avoid harm. This doesn't really indicate anything.

> this reflection reveals are the same experiences that predate our ability to self reflect

That's the key though. Self-awareness is the distinction.

> I guess what I’m fundamentally not understanding in your argument is the basis for the idea that a species gaining self reflection somehow becomes the point at which it becomes immoral to kill or harm that species.

Without self-awareness, they are not a 'person', essentially just more complex automata. They can't shape their environment, they are just a part of their environment, following their instincts.

They don't think, therefore they are not.

> Furthermore, moral behavior can be found all throughout the animal world, with clear indications of love/protection, companionship, sharing/cooperation, reciprocity, memory of transgressors, etc.

Some of that is just programmed instinct, quite different from humans. For examples, some mothers will attack their young, does this mean they 'love' them, or they have a programmed instinct to protect their young? Some of those same mothers will eat some of their young also, keep in mind before you answer.

> I’d argue that gaining the ability to self reflect is the very thing that increases our moral obligations.

Sure, to reduce harm, but not to not take a life.

Take a cod for example. It has no self-awareness, no personhood, no traits worth valuing. Its body is worth more than it's life, and if killed humanely no harm is done.

> the implications of it mattering are worth exploring more broadly.

I agree. But I've spent the last few years debating and researching this stuff, and I've come to my conclusions. They are in line with our current scientific understanding, and unless something changes it's what will continue to make sense to me.

Post reply on HN