[ExI] Dr. GPT, Problem-solver

William Flynn Wallace foozler83 at gmail.com
Mon May 1 16:05:28 UTC 2023


Yea for Ben!  Let's get real:

Life.  Consciousness.  Intelligence. etc.

Things hard to define.  What good will come when we agree on the meanings
of these words?   Is an AI conscious or not?  Who cares?  Why should we
care?

A biologist I told once that I had read a book called "Life on Man" said
that the author's definition of a parasite was different from his.  And I
said "Well, that doesn't make either one of you right or wrong, does it?"

He had no reply, of course.  Biologists will deal with the creatures
whatever we call them, won't they ?  Yep.  Will deciding that a slime mold
is intelligent cause us to do anything different?  Nope.  Don't argue with
words:  argue with observations of the slime mold.

We start out life putting things in our environment into one of two
categories:  to be feared, or not to be feared.  Thus begins our journey of
putting people and things in our lives into the 'right category'.  "Is she
cheating or not?"  "Am I doing the right thing?"  Much of this makes a
difference.

Much doesn't.  Who cares?

I suggest that we should treat definitions/categorizations just like we do
theories: we don't ask that they be right or wrong.  We ask that they be
useful. To be useful they have to produce testable ideas which can be
falsified.

Is something intelligent has to depend on defining intelligence in ways
that leave no doubt as to what we are talking about/ testing.

So stop the word games and deal with scientific inquiry - data.
Interpretations will come later and be discussed and argued endlessly, as
we know.

Ya know, some days it seems to me that my life, inner and outer, my
environments and so on, are being directed by an evil demon.  Nothing else
makes any sense at all.   bill w


On Mon, May 1, 2023 at 10:40 AM Adrian Tymes via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On Mon, May 1, 2023 at 1:33 AM Ben Zaiboc via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
>> I have an idea.
>>
>> Instead of endlessly arguing and giving these GPT systems leading
>> questions about whether they are conscious or understand things then
>> believing their answers when they correspond to our preconceptions, why
>> doesn't somebody ask one how it can help us cure cancer?
>>
>> Or any number of really important real-world problems.
>>
>> I mean, who cares if it 'really understands', when the real question is
>> can it really solve real problems?
>>
>
> Alas, it can't.  Not that one, at least.
>
> How might all forms of cancer be cured with a single, universally
> applicable solution?
>
> Unfortunately, it is unlikely that there will be a single, universally
> applicable solution to cure all forms of cancer. Cancer is not a single
> disease, but rather a collection of diseases that arise from different
> types of cells and genetic mutations. These different types of cancer
> require different treatments, and there is no one-size-fits-all approach
> that can be applied to all types of cancer.
>
> Current cancer treatments, such as chemotherapy, radiation therapy,
> surgery, and targeted therapy, are highly specific to the type of cancer
> being treated. Even within a single type of cancer, different patients may
> require different treatments based on the unique genetic makeup of their
> tumors. Therefore, developing a single solution that can cure all types of
> cancer is highly unlikely.
>
> That being said, there are some promising areas of research that may lead
> to more effective and targeted cancer treatments. For example,
> immunotherapy, which harnesses the power of the immune system to fight
> cancer, has shown promising results in some types of cancer. Similarly,
> precision medicine, which tailors treatments to the individual genetic
> profile of a patient's tumor, has the potential to improve outcomes for
> cancer patients.
>
> Overall, while it is unlikely that there will be a single, universally
> applicable solution to cure all forms of cancer, continued research and
> development of targeted and personalized therapies may lead to more
> effective treatments for individual types of cancer.
> Regenerate response
>
>
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20230501/8e5f7e8a/attachment-0001.htm>


More information about the extropy-chat mailing list