If the brain is just bioelectrical hardware then there is nothing worth reverse engineering, nor is there any worth to anything at all.

You can simulate digestion with a turing machine all you want, but that simulation wont digest any food. Likewise you can simulate the neurons in a brain but that simulation can not be an instantiation of the concept of triangularity.

Of course there's something worth reverse engineering, because in that case doing so would allow us to create consciousness as well as transfer and copy it. It would hold the key to at least a form of practical immortality.

If the brain is not just bioelectrical hardware, then we have the problem of whatever it might "interface with" can be influenced and copied, and we don't even know if there is anything to interface with and copy.

> You can simulate digestion with a turing machine all you want, but that simulation wont digest any food. Likewise you can simulate the neurons in a brain but that simulation can not be an instantiation of the concept of triangularity.

This is an utterly meaningless paragraph. If we build a robot we most certainly could build one that actually digests food. To my knowledge we can't yet do so without actually losing energy rather than gaining it, but that is an issue with the efficiency of the engineering, no fundamental blocker. If in software, it wouldn't digest any food, sure, but it also would have no need to, and that has no parallel to the second part of your paragraph.

The "concept of triangularity" can certainly be simulated in an abstract manner. It doesn't even take a complex system. AI doing symbolic manipulation of abstract concepts is one of the earliest things we tried. It's not the hard part. The hard part is actually reasoning about those concepts, inferring them, being able to act on them and talk about them. But that too is orthogonal to solving the issue of what consciousness actually is.

Afaik we can throw bacteria at it and then use something similar to fuel cells to digest their products. There are already robots that you can feed by placing apple slices at the right location on them. Methane fuel cells driven by methane from anaerobic bacteria is not hard. The digestion is used for a long time in waste water treatment, and the fuel cells are good enough to be useful for a few years now. It's just that those fuel cells are _expensive_, and the system power density is quite meager, all considered. It also lacks an immune system that keeps the bacteria in check, so it's not that hard to poison (still quite resilient though).

No, it would not allow you to create consciousness, it would simply mean that there is no such thing as consciousness. eliminative materialism would be true. there would be no 'you' to live forever.

Us not knowing exactly how something works has never stopped us before from looking into it.

With "a simulation of digestion" I specifically mean a software simulation, not a hardware simulation. I assure you I had a meaning with the text I wrote, it is not meaningless. The meaning is for me to say, your interpretation of it (and this is the point) underdetermines the meaning of it. You fully determine what i meant simply from the characters on the screen. Physical instantiations of abstract items work the same way. An item needing to do something is irrelevant of it being able to, a software simulation of food does not actually digest food. A software simulation of abstract concepts do not actually instantiate them.

I am repeating myself again but any AI system you believe to have instantiated the abstract concept of triangularity is just an interpretation on your part to having done so, you might as well interpret it to simply being a room heating device or the implementation of some other algorithm that also matches the behaviour.

> No, it would not allow you to create consciousness, it would simply mean that there is no such thing as consciousness. eliminative materialism would be true. there would be no 'you' to live forever.

Do you believe you are conscious? I believe I am conscious. If the brain is purely bioelectrical, then a purely bioelectrical system at least is able to reproduce an entity that believes itself to be conscious.

Whether there is a meaningful distinction between that and "actual" consciousness is something we simply don't know, because we don't have a way of measuring whether or not another entity actually is conscious, whatever we take that to mean.

Maybe that belief that I am conscious is flawed and merely an illusion, but if copying my mind could allow that illusion to persist forever, then my illusion of consciousness would still be quite pleased with that outcome. That is quite enough for me.

> I am repeating myself again but any AI system you believe to have instantiated the abstract concept of triangularity is just an interpretation on your part to having done so, you might as well interpret it to simply being a room heating device or the implementation of some other algorithm that also matches the behaviour.

Here is an implementation of the abstract concept of triangularity that does not involve an interpretation of what it means or an instantiation of the concept, merely a symbol that describes the abstract concept as it is:

    thought = :triangularity

I define the symbol :triangularity to refer to the abstract concept, so it does. I can manipulate that concept with or without making any interpretation as to what the concept actually means.

Symbolic manipulation of abstract ideas is trivially easy. The idea that we can't is ludicrous. The challenge comes when we try to interpret and reason about specific concrete interpretations, not in holding an abstract concept - we've known how to communicate abstract concepts as data without instantiating an interpretation since at least the moment we could write.

You're attacking AI on the basis of one of the problems that are easiest to solve. Symbolic manipulation of abstract concepts with computers in a meaningful form dates back to at least the 50's. The ease with which we can do that was part of much of the early misguided hope of a rapid revolution in general artificial intelligence.

> Maybe that belief that I am conscious is flawed and merely an illusion

That's the Dennett nonsense. Of course our sense inputs are an illusion, it's that there is anything being perceived at all that is the mystery! :

DENNETT: So you see, Socrates, consciousness is an illusion.

SOCRATES: I see, but I need to clarify something. What do you mean by illusion?

DENNETT: An illusion is an appearance contrary to reality.

SOCRATES: And what is an appearance?

DENNETT: An appearance is something that is apprehended by an observer.

SOCRATES: So consciousness is something that is apprehended by an observer which is contrary to reality.

DENNETT: But that cannot be, because then the observer would have to be conscious.

SOCRATES: Therefore, consciousness cannot be an illusion!

This makes no sense to me.

The point I was making is that we do not know what consciousness actually is. Until we do, we have no way of telling whether my impression of my own consciousness has any reasonable relation to the reality of it. It is meaningless to try to determine whether or not it is an illusion without first having an understanding of what consciousness is.

More directly to the argument I was addressing, the point is that whether or not the commenter I replied to is right and there'd be no "me" that could live on forever, that is really rather irrelevant to the "me" that believes I am conscious.

In other words: While I need clarity in what consciousness is to be able to even begin to address whether or not what I have is truly consciousness, it really doesn't matter to the question of whether or not whatever I am sees value in a process that'd let me live on.

You don't have an impression of your own consciousness (your conscious awareness) to begin with. You just perceive stuff. How exactly could you not have conscious awareness?

We can't even meaningfully answer that question without an understanding of what consciousness is first.

Consciousness appears to us as if we are distinct entities, with a spatial and temporal existence, operating on sensory input and memory. We don't know if any of that is true - we don't know if there is any persistent entity, whether spatially or temporally.

I'm not saying there's "nothing" that we could label consciousness. I'm saying we have no way of suggesting whether or not our idea of ourselves as an entity that experiences events and acts on them (even just mentally) is meaningfully connected to reality.

My point in making that argument is that it doesn't matter - absent any better information, the only thing that makes sense for us is to act on the basis that our consciousness is real, and interact with people on the basis that they're real and conscious, because it doesn't matter whether or not it is an illusion if we can not "see past the veil".

This is incidentally also the classical response to philosophical idealism: Sure, maybe all our senses lie to us, but since everything acts as if we're in a materialist world, it doesn't matter.

It remains possible that at some point we find something that lets us lift the veil, and peer out into something "outside" our present reality. But that possibility does not change anything until and unless it becomes reality.

It does not follow that " a purely bioelectrical system at least is able to reproduce an entity that believes itself to be conscious" simply because you believe yourself to be conscious. A human simply being purely bioelectrical is the very thing in question. Further, a belief, as somethign abstract, is not something that a bioelectrical system could in principle capture, and that we do know, in such a scenario there is no 'you' to be 'pleased' any more than a rock could be.

"thought = :triangularity"

That is simply some LCD pixels on my screen. What you define it to be has absolutely no bearing on the physical implementation of the pixels. You could likewise define it to be a partial sentence on 3d polygonal mesh generation from a math phd thesis or any number of other interpretations. Any interpretation you make is IN YOUR HEAD. Not in the physical instantiation (the LCD pixels or memory the transistors of the RAM).

The problem here is that you don't understand the point.

> It does not follow that " a purely bioelectrical system at least is able to reproduce an entity that believes itself to be conscious" simply because you believe yourself to be conscious

You are right, that does not follow. It does however follow that if we are purely bioelectric systems, and that is sufficient for us to believe ourselves to be conscious, then it is reasonable to presume that a clone of such a system will presume itself to be conscious. That is the claim I was making.

> A human simply being purely bioelectrical is the very thing in question.

You cut of "If the brain is purely bioelectrical" from the part you quoted above, and so what you attacked above was a strawman of your own making, and not the argument I made.

> That is simply some LCD pixels on my screen. What you define it to be has absolutely no bearing on the physical implementation of the pixels.

That is entirely irrelevant. It is a representation of an implementation of an symbolic representation of an abstract thought in a programming language that runs on a Turing machine (it is valid Ruby). As such it is proof that a turing machine can represent abstract thought.

> The problem here is that you don't understand the point.

We clearly do not understand each other, but so far I have seen nothing to suggest that you have a meaningful argument. You're trying to create a distinction between the abstract and an instantiation of the idea to create a separation that makes representing the abstract notion in a computational device impossible. But representing abstract thought in symbolic form is something humanity has been capable of four hundreds if not thousands of years at a minimum. The point is that abstraction and symbolic representation does not get any less real because you invent additional intermediate steps.

Sure, what you read appeared to you as some LCD pixels on your screen, but that does not make it any less a symbolic representation of the abstract idea, rather than an interpretation of the idea. If you fail to understand it and instead interpret it as a sentence on 3d polygonal mesh generation is irrelevant - it does not change the abstract notion, any more than misinterpreting a thought changes the original thought.

You argued we could not represent anything but an interpretation of an abstract idea. The above is proof this is not true, because any abstract idea can be exchanged for a symbol representing that idea, and the symbol can be accurately reproduced.

This is a foundational aspect of communications, of symbolic logic, of linguistics, and a whole range of other scientific disciplines, because it is such a central aspect of the ability for human discourse about abstract ideas.

If you want to attack AI there are many angles to take - we do not know whether or not consciousness requires some form of dualism for example, hence my constant qualifications ("if the brain is purely bio-electrical") about it. That's a weak spot.

But our ability to represent abstract ideas in computation is not one of the weak spots. It's one of the easiest parts of AI, or indeed of any computation.

The understanding of how to represent abstract notion by symbolic substitution is old enough that you see that transition happening in language going back thousands of years, when concrete concepts starts being used as symbols of abstract ideas.

That does not follow either for if we are purely bio-electrical then we can not have beliefs.

Strawman? What?

Programming languages running on turing machines are abstract entities, they only exist in your mind, not in reality. The physical machine just shuffles electrons around.

Abstract Thought -> Symbolic Representation -> Physical Implementation.

The right hand side is not the left hand side. Any right hand side implements an infinite left hand sides and there is no way for you to pick out the correct one simply from the physics. I have never written the word representation before in this exchange, yet you claim i have. Stop doing that.

You wont get it. This is a standard part of philosophy of mind. Look it up.

> If the brain is just bioelectrical hardware then there is nothing worth reverse engineering, nor is there any worth to anything at all.

may I introduce you to our local nihilist club

If we can make a faster or parallel perfect copy that would be rather valuable - uploading, copying and "fork/join" of minds would be rather useful.

not nihilism, materalistic eliminativism - just happens to have nihilism as a consequence.

> You can simulate digestion with a turing machine all you want, but that simulation wont digest any food.

You could digest simulated food.

You could also build a machine stomach to digest real food.

If you found out you were a simulation you’d be looking pretty dumb right now

> then there is nothing worth reverse engineering, nor is there any worth to anything at all.

Non-sequitor. You are certain to die, so there's no point in continued breathing?

It is not a non-sequitor, it follows from the concepts, something being worth doing is not a part of physics/math. If math/physics is all you have then you have no "worth doing".

If it turns out we're just simulated entities, does that suddenly make your life not worth living?

Why would it change anything?

if we are simulated entities then there is no "we". 'we' do not have existence. that would indeed make life not worth living, on account of there not being a life there in the first place

If we are simulated entities, then the fact that I at least live a pleasant life would show that it is perfectly possible to have a pleasant, worthwhile simulated life.

no it would show that you are not simulated

It might, if it turns out there's an opportunity for a reset.

> You can simulate digestion with a turing machine all you want, but that simulation wont digest any food.

Because food is a physical item. Ideas and abstract concepts are not and can be communicated in a format that's fully compatible with a simulated consciousness. There is no reason that wouldn't be possible aside from mechanisms that are, or at least appear, supernatural.