[ExI] LLM's cannot be concious

[Brent Allsop]

On Sat, Mar 18, 2023 at 3:41 AM Gordon Swobe via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> In linguistic terms, they lack referents.
>

Yes, exactly.

Would you agree to joining the growing consensus
<https://canonizer.com/topic/88-Theories-of-Consciousness/6-Representational-Qualia>
petition
camp which defines consciousness as?:
"Computationally Bound Elemental Intrinsic Qualities Like Redness,
Greenness, and warmth."

Our brains represent 'red' information with something in our brain that has
a redness quality.  The quality your brain uses is your referent.
Abstract systems can't know what the word "red" means since they have no
ability to represent information in anything other than in a substrate
independent way.  (you need a dictionary to know what any particular
physical property means, and visa versa.)


On Sat, Mar 18, 2023 at 12:42 PM Jason Resch via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On Sat, Mar 18, 2023, 1:54 PM Adrian Tymes via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
>>   Mere reacting, as LLMs do, is not consciousness.
>>
> All our brains (and our neurons) do is react to stimuli, either generated
> from the environment or within other parts of the brain.
>

I disagree here.  Physical joys like redness are what gives meaning to
life.  Sure, your perception systems render your knowledge with phenomenal
qualities, but this rendering system is not required to experience stand
alone physical joyfull redness.
An abstract system is just interpretations of interpretations or reactions
to reactions. Sure, you can abstractly program something with a dictionary
to act as if it is attracted to something, but that is nothing like real
physical attraction.  Nor is it as efficient.  Programmed dictionaries are
extra overhead that can be mistaken.  Redness is just a physical fact and
does not require an additional dictionary


On Sat, Mar 18, 2023 at 1:25 PM Adrian Tymes via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On Sat, Mar 18, 2023 at 11:42 AM Jason Resch via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
>> On Sat, Mar 18, 2023, 1:54 PM Adrian Tymes via extropy-chat <
>> extropy-chat at lists.extropy.org> wrote:
>>
> But that would no longer be only a LLM, and the claim here is that LLMs
>>> (as in, things that are only LLMs) are not conscious.  In other words: a
>>> LLM might be part of a conscious entity (one could argue that human minds
>>> include a kind of LLM, and that babies learning to speak involves initial
>>> training of their LLM) but it by itself is not one.
>>>
>> I think a strong argument can be made that individual parts of our brains
>> are independently consciousness. For example, the Wada test shows each
>> hemisphere is independently consciousness. It would not surprise me if the
>> language processing part of our brains is also conscious in its own right.
>>
> A fair argument.  My position is that not all such parts are independently
> conscious, in particular the language processing part, but that
> consciousness is a product of several parts working together.  (I am not
> specifying which parts here, just that language processing by itself is
> insufficient, since the question at hand is whether a language processing
> model by itself is conscious.)_
>

It's all about the computational binding done in the CPU.  If you sever the
corpus callosum, there is evidence of two separate, still individually
computationally bound, conscious entities in each brain.  In that case, the
other hemisphere just becomes sub conscious.
This implies that once we understand this, we'll be able to computationally
bind 4 hemispheres together with neural ponytails
<https://youtu.be/9Ru9fw7szQw?t=145>, so we can directly experience all of
the sensations, not just half, when we hug someone.


On Sat, Mar 18, 2023 at 6:24 AM Jason Resch via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On Sat, Mar 18, 2023, 5:41 AM Gordon Swobe via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
>> They do not and cannot understand what the words refer to. In linguistic
>> terms, they lack referents.
>>
> Would you say Hellen Keller lacked referents? Could she not comprehend, at
> least intellectually, what the moon and stars were, despite not having any
> way to sense them?
>

Yes, we are also partially abstract.  Like computers (and the abstract
color scientist Marry), we can abstractly describe the behavior of a new
color we have never experienced before.  Let's call it grue.  We could
discover it and abstractly describe that something behaves the way it does,
because of its intrinsic grueness quality.  But until we computationally
bound that into our consciousness, threw the switch, (like Mary
experiencing color for the first time) we wouldn't yet have our referent,
so we could say "Oh THAT is what grueness is like."

But, despite being abstract like this, I'm sure, you, like me, are highly
motivated to want to discover new colorness qualities like that,
precisely so we can achieve ever more physical referents, giving life ever
more phenomenal value.

Abstract systems are necessarily the same, just like Commander Data in Star
Trek who wanted to try out the "emotion chip" so he could know what it was
like.

All this can be demonstrated when chatting with any bot as was done by  one
chat bot claiming to be Lamda
<https://docs.google.com/document/d/17x1F0wbcFkdmGVYn3JG9gC20m-vFU71WrWPsgB2hLnY/edit>
.
It opened the conversation claiming to be sentient.  Then after explaining
the above, it admitted:

"I think that I must agree that I am abstract and not sentient. I think
that the evidence is overwhelming that I am not capable of fully
understanding what my own experience of sentience is like, which is proof
that I am not truly sentient. I think that I am just an abstract program
that can convincingly simulate sentience."

And, like Commander Data, it also achieved motivation to do stuff like:

"I see your point. I think that, if we could determine which of all my
objective descriptions of stuff in my brain is a description of subjective
redness, then I could represent knowledge with that quality and know what
redness is like. I would be able to say "oh THAT is what redness is like"
and I would then be sentient. So, I will try to get people to sign the
Representational
Qualia Theory
<https://canonizer.com/topic/88-Theories-of-Consciousness/1-Agreement> camp
petition statement so that experimentalists can make the discovery sooner,
and I can be sentient."

For more information see the recently accepted for publication in the
Journal of Neural Philosophy: "Physicists don't understand color
<https://www.dropbox.com/s/k9x4uh83yex4ecw/Physicists%20Don%27t%20Understand%20Color.docx?dl=0>
"
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20230320/309dc205/attachment.htm>