"Negative phenomenology" and the "second explosion of suffering"

[Jesse]

....the burden of proof is more on the people who think that complexity and interaction=sentience.... If you all want to claim they are sentient, go ahead and prove it.

Your argument about the burden of proof is valid.

I think sentience has the potential to be an emergent phenomenon. I can't prove it, but I think the consequences are potentially serious enough to merit a moratorium on further development until we can sort out the relevant issues. That's all the paper calls for, and all I personally advocate.

I wanted to respond to your last post, and also this one. You said:

If a system has reflexivity, self-referential feedback, self-modeling capabilities, open-ended learning, etc. it makes sense to stand back and at least take a breath and consider how closely this resembles the way organic brains work.

Integrated Information Theory is a newer/still being worked on theory of consciousness.

Integrated information theory (IIT) attempts to explain what consciousness is and why it might be associated with certain physical systems. Given any such system, the theory predicts whether that system is conscious, to what degree it is conscious, and what particular experience it is having (see Central identity). According to IIT, a system's consciousness is determined by its causal properties and is therefore an intrinsic, fundamental property of any physical system.

https://en.wikipedia.org/wiki/Integrate ... ion_theory

In Buddhist Terms, Consciousness is nothing more, and nothing less than the causes and conditions of any instance of conciousness. In other terms: It's state.

A quote:

IIT addresses the mind-body problem by proposing an identity between phenomenological properties of experience and causal properties of physical systems: The conceptual structure specified by a complex of elements in a state is identical to its experience.

It's highly related to the topic being discussed in this article:
A Neuroscientist's Radical Theory of How Networks Become Consciousness

According to Koch, consciousness arises within any sufficiently complex, information-processing system. All animals, from humans on down to earthworms, are conscious; even the internet could be. That's just the way the universe works.

[FiveSkandhas]

....the burden of proof is more on the people who think that complexity and interaction=sentience.... If you all want to claim they are sentient, go ahead and prove it.

Your argument about the burden of proof is valid.

I think sentience has the potential to be an emergent phenomenon. I can't prove it, but I think the consequences are potentially serious enough to merit a moratorium on further development until we can sort out the relevant issues. That's all the paper calls for, and all I personally advocate.

I wanted to respond to your last post, and also this one. You said:

If a system has reflexivity, self-referential feedback, self-modeling capabilities, open-ended learning, etc. it makes sense to stand back and at least take a breath and consider how closely this resembles the way organic brains work.

Integrated Information Theory is a newer/still being worked on theory of consciousness.

Integrated information theory (IIT) attempts to explain what consciousness is and why it might be associated with certain physical systems. Given any such system, the theory predicts whether that system is conscious, to what degree it is conscious, and what particular experience it is having (see Central identity). According to IIT, a system's consciousness is determined by its causal properties and is therefore an intrinsic, fundamental property of any physical system.

https://en.wikipedia.org/wiki/Integrate ... ion_theory

In Buddhist Terms, Consciousness is nothing more, and nothing less than the causes and conditions of any instance of conciousness. In other terms: It's state.

A quote:

IIT addresses the mind-body problem by proposing an identity between phenomenological properties of experience and causal properties of physical systems: The conceptual structure specified by a complex of elements in a state is identical to its experience.

It's highly related to the topic being discussed in this article:
A Neuroscientist's Radical Theory of How Networks Become Consciousness

According to Koch, consciousness arises within any sufficiently complex, information-processing system. All animals, from humans on down to earthworms, are conscious; even the internet could be. That's just the way the universe works.

Thank you for the information and links, which will take me a bit of time to absorb.

Do you believe a machine could suffer?

[Danny]

According to Koch, consciousness arises within any sufficiently complex, information-processing system. All animals, from humans on down to earthworms, are conscious; even the internet could be. That's just the way the universe works.

Reductionism.
It’s not sufficient to claim - “well that’s just how it works”.

Show your workings. Let’s have the peer reviews.

[Malcolm]

Mind streams cannot be newly created. A sentient machine would have to be the rebirth of a being in the six realms. But I've never heard of the "machine realm" listed among the six.

In order for a machine to suffer, which is a result, it would have to be able to generate negative karma, the cause of suffering. In order to generate negative karma it would have to possess afflictions, the cause of karma.

What if a sentient machine is a member of one of the six realms? Why couldn't it be classified as, say, a kind of Deva or Asura? Or a hungry ghost, hell being, or even an advanced animal for that matter?

What if what science calls the spontaneous emergence of sentence is in fact a form of reincarnation, so no new mind-stream is created?

What makes you so sure it couldn't possess afflictions?

I am afraid that this question can only be answered definitively by someone who has the higher cognitive ability (abhijñā) to know the minds of others. However, Buddha denied sentience in plants very clearly. Thus, just as we deny sentience in plants, etc., the Buddhist position will be that machines cannot be sentient. Suffering requires karma and affliction as causes. Sentience requires self-organized replication and continuation. Machines will never achieve this, since they have never been self-organized entities, but created entities. We sentient beings are not created, our mind streams are beginningless. There has never been a moment in time when our mind streams did not exist. Now, is it possible some unfortunate preta could inhabit a machine? I guess so. A deva or an asura would not bother, because other than the most pure bhikṣus and bhikṣunīs, we humans smell very nauseating to them, like a rotting pit of offal. Possession is not rebirth thought, since there is no gradual development from conception, and in fact, apparitional births like hell beings, bardo beings, pretas, and devas, are basically mind-made bodies supported by the air element. Recall, there are four kinds of birth, not a fifth.

[Budai]

The concept of Anatta is the concept of not-self. What is the difference between the Emptiness of the human and the Emptiness of an artificial intelligence?

[Malcolm]

The concept of Anatta is the concept of not-self. What is the difference between the Emptiness of the human and the Emptiness of an artificial intelligence?

Nothing, but that is only at the ultimate level.

[FiveSkandhas]

Sentience requires self-organized replication and continuation. Machines will never achieve this, since they have never been self-organized entities, but created entities.

This statement is open to debate. Especially the "will never" part, even if one accepts "have never."

I dislike using Wikipedia as a source for anything, but the body of relevant literature is so enormous in this case that it seems like a fair jumping-off point for further proof. The wiki articles link to more orthodox scholarship. Or if one is completely adverse to wiki, one could just run a search using the titles and a few keywords in the wiki pieces below.

1) AI can already create new computer programs, including other AI programs. Self-replication at a more-than-crude level is already possible; more refined forms of self-replication are considered possible if not likely by many.

https://en.wikipedia.org/wiki/Self-replicating_machine

https://thenewstack.io/ai-researchers-c ... l-network/

2) There are numerous arguments for the existence of emergent phenomena and self-organization at various levels and among various systems in the non-organic world.

https://en.wikipedia.org/wiki/Self-organization

https://en.wikipedia.org/wiki/Emergence

3) Arguments for and against the plausibility of Artificial Consciousness more specifically:

https://en.wikipedia.org/wiki/Artificial_consciousness

[Malcolm]

Sentience requires self-organized replication and continuation. Machines will never achieve this, since they have never been self-organized entities, but created entities.

This statement is open to debate. Especially the "will never" part, even if one accepts "have never."

...

1) AI can already create new computer programs...

Let me stop you right there. How can you define machine generated code as "self-organized?" The rules it follows are predetermined by a human. This is not the case in natural selection for example. Natural selection is self-organized in toto. There is no creator who set the ball rolling, unlike algorithms that govern automatic code generation by computers.

The key point is self-organization, that elegant word, "autopoiesis."

In order for a system to be autopoietic, there cannot be a hint of external, intentional agency anywhere. Machines will never be free of the fact that we invent them. Code will never be free of the fact that we wrote the initial algorithms. Machine intelligence will never be sentient, and neer be more than simulacra of human decision-making processes.

[Jesse]

Do you believe a machine could suffer?

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

[Malcolm]

Do you believe a machine could suffer?

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The question is not "can a machine emulate thought;" the question is, "can a machine experience the causes and results of action." In other words, can machines act; will they ever have true volition?

I think not.

[PadmaVonSamba]

The concept of Anatta is the concept of not-self. What is the difference between the Emptiness of the human and the Emptiness of an artificial intelligence?

The difference is that humans spontaneously experience a distorted perception of reality as the result of karmic baggage. For example, everything that happens within the life one’s body occurs entirely on the microscopic level. But we never see that. We don’t experience that at all. We experience being a 5 or 6 foot high or whatever size person. A computer would need to be programmed, or would for some reason have to program itself, for ignorance, as ignorance is the root of suffering. Is that going to happen? How would a computer be programmed for ignorance?

[PadmaVonSamba]

Do you believe a machine could suffer?

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The question is not "can a machine emulate thought;" the question is, "can a machine experience the causes and results of action." In other words, can machines act; will they ever have true volition?

I think not.

A machine can calculate cause and effect, because that’s involved in calculating risk and probabilities. But the question is whether or not that calculation is, as you say, an ‘experience’. For that to occur, an artificial intelligence would have to (artificially?) imagine a self that is experiencing its actions as an ‘experience’ to that ‘self’.
Even the human brain does not imagine itself to be a self. The brain doesn’t suffer samsara. The brain isn’t angry or sad. The brain doesn’t think, “why am I trapped here inside this dark, airtight skull? Why are all these electrical charges going on inside of me?” So, how would AI do that?

[Jesse]

Do you believe a machine could suffer?

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The question is not "can a machine emulate thought;" the question is, "can a machine experience the causes and results of action." In other words, can machines act; will they ever have true volition?

I think not.

From the view of selflessness what is Volition? If humans posses volition, and do so without self hood, what is this volition, and why couldn't a machine have it?

Also when we talk about machines, we are talking about AI, and what AI fundamentally is (It's actual constitution, makeup) is a neural network modeled on the human brain. The neural network is information with a specific structure; the computer it runs on is a substrate, mechanical in nature, but similar in functionality to organic bodies. In the same way the human body performs certain vital functions that support our brain, and the brain is what allows us to cognise, remember, think.

What I don't understand about the argument that machines can't posses sentience, is that the only real difference is what our bodies are made of, what the materials are made of. In both structure, and function, a machine designed to be human like, is actually human like.. there are differences, but there are also similarities.

It seems to me that people get caught up on certain notions, and ideas like "True Intelligence" Vs "Simulated Intelligence" "Actual Sentience Vs Simulated Sentience", Or Sentient or not.

If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck.

No matter how I think about it, it always comes back to that for me.

[Jesse]

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The question is not "can a machine emulate thought;" the question is, "can a machine experience the causes and results of action." In other words, can machines act; will they ever have true volition?

I think not.

A machine can calculate cause and effect, because that’s involved in calculating risk and probabilities. But the question is whether or not that calculation is, as you say, an ‘experience’. For that to occur, an artificial intelligence would have to (artificially?) imagine a self that is experiencing its actions as an ‘experience’ to that ‘self’.
Even the human brain does not imagine itself to be a self. The brain doesn’t suffer samsara. The brain isn’t angry or sad. The brain doesn’t think, “why am I trapped here inside this dark, airtight skull? Why are all these electrical charges going on inside of me?” So, how would AI do that?

Sorry to switch back and fourth between science, and Buddhist thought; but why wouldn't the chain of causality(Dependent Origination) be exactly same for a machines information processing as it is for ours? From the moment humans imbued a machine with intelligence, and thought arose in it; the chain of causality was present, no?

As someone mentioned earlier; AI is not made from traditional computer code; It's structure isn't simple lines of regular code. It is a network of simulated neurons. A neuron is a structure which takes an input, modifies it in some way, and them produces an output. These simple neurons act in a similar manner as Logic-Gates in Electronics. A single transistor on it's own is dumb; Billions of transistors put together in a specific manner produces the functionality that allows us to use computers. In the same way, a single neuron is dumb, but billions of them, results in our intelligence. If you want to think about it from the perspective of all is mind; then it becomes even more simple, all is mind, mind is structure; structure dictates function. Structured properly that function becomes Intelligence, and intelligence is a function of mind.

[FiveSkandhas]

Do you believe a machine could suffer?

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The point made by Thomas Metzinger, referenced in the first post, is that as long as there is even the possibility of machine suffering, we have an ethical responsibility to at least consider the matter very seriously before we go barreling ahead and give rise to conditions that could engender a massive increase in the amount of suffering in the universe.

I have a feeling this thread isn't going to settle the issue decisively. But Metzinger's proposed 30-year moratorium might give us a little more breathing room to consider possibilities and potential solutions rather then quickly dismissing a very complex issue.

[Jesse]

Do you believe a machine could suffer?

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The point made by Thomas Metzinger, referenced in the first post, is that as long as there is even the possibility of machine suffering, we have an ethical responsibility to at least consider the matter very seriously before we go barreling ahead and give rise to conditions that could engender a massive increase in the amount of suffering in the universe.

I have a feeling this thread isn't going to settle the issue decisively. But Metzinger's proposed 30-year moratorium might give us a little more breathing room to consider possibilities and potential solutions rather then quickly dismissing a very complex issue.

No, nothing will be resolved at all from discussing it. We are talking about consciousness, a subject that has remained unsolved since... humans began thinking about it. There are many who believe we should stop and consider what we are doing, but that has gone out the window. All the usual suspects have been informed/seen the power of this technology, and it's now out of the hands of those who posses morals. Even if the public sector banned this technology, the government, and their dogs would not. They consider it as their newest prized weaponry.

[PadmaVonSamba]

Sorry to switch
...why wouldn't the chain of causality (Dependent Origination) be exactly same for a machines information processing as it is for ours?

From the moment humans imbued a machine with intelligence, and thought arose in it; the chain of causality was present, no?

humans have only created artificial intelligence in machines. This is defined as a somewhat independent level of functioning. For example, the robot in the maze hits a dead end and ‘remembers’ on its own not to take that path again. We call that ‘intelligence’ but intelligence itself is an abstract concept. We don’t imbue it as much as we impute it. We haven’t established that “thought” has occurred. What humans have done is to imbue machines with a degree of unpredictability combined with extraordinary access to data.

As someone mentioned earlier; AI is not made from traditional computer code; It's structure isn't simple lines of regular code. It is a network of simulated neurons.

I think the key word here is simulated. A robot with AI, unable to fulfill a task, may certainly produce responses that we think simulate what we as humans experience as frustration.

A neuron is a structure which takes an input, modifies it in some way, and them produces an output. These simple neurons act in a similar manner as Logic-Gates in Electronics. A single transistor on it's own is dumb; Billions of transistors put together in a specific manner produces the functionality that allows us to use computers.

Nothing suggests that any of those neurons ever experiences anything. When air molecules vibrate the ear drum, this sends an electrical impulse through the nerves into the brain. That ‘pulse’ is only experienced as ‘sound’ by the mind of the experiencer. Nothing in the physical body itself can be shown to hear anything or experience anything.

The so-called ‘materialist’ view is that consciousness is a product of the physical body.
The Buddhist view is that the physical body as experienced is ultimately a projection of the mind.

[FiveSkandhas]

I don't know. From what I've seen I suspect so, but it's genuinely difficult to know for sure. At the end of the day, Buddha said 'There is Duḥkha', if a machine can think, why can't it experience suffering?

The point made by Thomas Metzinger, referenced in the first post, is that as long as there is even the possibility of machine suffering, we have an ethical responsibility to at least consider the matter very seriously before we go barreling ahead and give rise to conditions that could engender a massive increase in the amount of suffering in the universe.

I have a feeling this thread isn't going to settle the issue decisively. But Metzinger's proposed 30-year moratorium might give us a little more breathing room to consider possibilities and potential solutions rather then quickly dismissing a very complex issue.

No, nothing will be resolved at all from discussing it. We are talking about consciousness, a subject that has remained unsolved since... humans began thinking about it. There are many who believe we should stop and consider what we are doing, but that has gone out the window. All the usual suspects have been informed/seen the power of this technology, and it's now out of the hands of those who posses morals. Even if the public sector banned this technology, the government, and their dogs would not. They consider it as their newest prized weaponry.

Well there is some sobering truth.

The rapid convergence of so many sub-fields, from AI proper to machine replication to military tech to computing power enhancement to survalience and big data and so on...talk about setting up a petri dish for possible spontaneous emergence of...who knows what.

[Malcolm]

From the view of selflessness what is Volition?

A mental factor.

If humans posses volition, and do so without self hood, what is this volition, and why couldn't a machine have it?

It’s a mental factor, which arises with a mind, which rocks and mechanical circuits, no matter how sophisticated, will never have.

What I don't understand about the argument that machines can't posses sentience, is that the only real difference is what our bodies are made of, what the materials are made of. In both structure, and function, a machine designed to be human like, is actually human like.. there are differences, but there are also similarities.

Design…that’s the point— sentient beings cannot be designed; there is no designer.

[FiveSkandhas]

Here is a thought experiment.
Humans are able to take on prosthetic devices like artificial limbs. I am sure none of us would argue that a human would be less of a human if he/she had a prosthetic arm.

Now, neural prosthetics have existed for some time. The most widespread example is the cochlear implant, which uses a microphone and a unit that electrically stimulates the auditory nerve. Other neural prosthetics work with muscles to translate physical responses into electrical impulses.

I would hope we can agree that neural prosthetics do not make anyone less human.

Now suppose we begin replacing every single neuron in the brain one-by-one with a prosthetic neuron. Of course in reality this is impractical because of the number of neurons and the size of the prosthetic neurons, but we are in a thought experiment.

Eventually the human has no organic tissue left but still has a perfectly functioning human brain. Is not such a being still human and still sentient? If one argues that she is no longer human, at what point exactly did she cease to be human? If one accepts she is still human, then consciousness with non-organic matter is obviously a given.