<div dir="auto">This is reminiscent of our recent debate:<div dir="auto"><br></div><div dir="auto"><a href="https://youtu.be/vjuQRCG_sUw">https://youtu.be/vjuQRCG_sUw</a><br></div><div dir="auto"><br></div><div dir="auto">Jason</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Apr 30, 2023, 6:37 PM Jason Resch <<a href="mailto:jasonresch@gmail.com">jasonresch@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><div><br><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Apr 30, 2023, 5:11 PM Gordon Swobe <<a href="mailto:gordon.swobe@gmail.com" target="_blank" rel="noreferrer">gordon.swobe@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto">The mere fact that an LLM can be programmed/conditioned by its developers to say it is or is not conscious should be evidence that it is not. </div></blockquote></div></div><div dir="auto"><br></div><div dir="auto">Should we take the ability of humans or animals to act or be trained as evidence they are not conscious?</div><div dir="auto"><br></div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><br></div><div dir="auto">Nobody wants to face the fact that the founders of OpenAI themselves insist that the only proper test of consciousness in an LLM would require that it be trained on material devoid of references to first person experience.</div></blockquote></div></div><div dir="auto"><br></div><div dir="auto">Their qualifications are as computer scientists, not philosophers of mind. Neither linguists nor AI researchers are experts in the field of consciousness. What does David Chalmers say about them? Have you looked?</div><div dir="auto"><br></div><div dir="auto">The test open AI proposes, it passed, would be strong evidence of human level reflexive consciousness. But failure to pass such a test is not evidence against consciousness. </div><div dir="auto"><br></div><div dir="auto">Also: Have you taken a few moments to consider how impossible the test they propose would be to implement in practice? Can they not think of an easier test? What is their definition of consciousness?</div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"> It is only because of that material in training corpus that LLMs can write so convincingly in the first person that they appear as conscious individuals and not merely as very capable calculators and language processors.</div></blockquote></div></div><div dir="auto"><br></div><div dir="auto">How do you define consciousness?</div><div dir="auto"><br></div><div dir="auto">Jason </div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><br></div><div dir="auto">-gts</div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Apr 30, 2023 at 7:30 AM Jason Resch via extropy-chat <<a href="mailto:extropy-chat@lists.extropy.org" rel="noreferrer noreferrer" target="_blank">extropy-chat@lists.extropy.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><div><br><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Apr 30, 2023, 5:23 AM Ben Zaiboc via extropy-chat <<a href="mailto:extropy-chat@lists.extropy.org" rel="noreferrer noreferrer" target="_blank">extropy-chat@lists.extropy.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
On 29/04/2023 23:35, Gordon Swobe wrote:
<blockquote type="cite">
<div lang="x-unicode">
<div><br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Sat, Apr 29, 2023 at
3:31 PM Ben Zaiboc via extropy-chat <<a href="mailto:extropy-chat@lists.extropy.org" rel="noreferrer noreferrer noreferrer" target="_blank">extropy-chat@lists.extropy.org</a>>
wrote:<br>
</div>
<div dir="ltr" class="gmail_attr"><br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="auto">So you believe them when they claim to not
be conscious, but don't believe them when they don't.<br>
<br>
And you expect us to take your reports of what they say
as evidence for whether they are conscious or not.<br>
<br>
Can you see a problem with that?</div>
</blockquote>
<div dir="auto"><br>
</div>
<div dir="auto">As I explained in another message, (to you,
I think), I first entered these discussions a couple of
months ago prepared to argue that people were being
deceived by the LLMs; that ChatGPT is lying when it says
it has consciousness and genuine emotions and so on.</div>
<div dir="auto"><br>
</div>
<div dir="auto">I had no personal experience with LLMs but a
friend had literally fallen in love with one, which I
found more than a little alarming.</div>
<div dir="auto"><br>
</div>
<div dir="auto">As it turns out, GPT4-4 is saying everything
I have always believed would be true of such applications
as LLMs. I’ve been saying it for decades.</div>
</div>
</div>
</div>
</blockquote>
<br>
Good grief, man, are you incapable of just answering a question?<br>
<br>
I suppose I'd better take your reply as a "No", you don't see a
problem with your double-standard approach to this issue.<br>
<br>
Please feel free to correct me, and change your (implied) answer to
"Yes".<br>
<br>
And when you say "prepared to argue...", I think you mean
"determined to argue...". But predetermined prejudicial opinions are
no basis for a rational argument, they are a good basis for a
food-fight, though, which is what we have here. One which you
started, and seem determined to finish.<br>
<br>
You may not have noticed (I suspect not), but most of us here
(myself included) have no dogmatic insistence on whether or not
these AI systems can or can't have consciousness, or understand what
they are saying. We are willing to listen to, and be guided by, the
available evidence, and change our minds accordingly. It's an
attitude that underlies something called the scientific method. Give
it a try, you might be surprised by how effective it is. But it
comes with a warning: It may take you out of your comfort zone,
which can be, well, uncomfortable. I suspect this is why it's not
more popular, despite how very effective it is.<br>
<br>
Personally, I think a little discomfort is worth it for the better
results, when trying to figure out how the world works, but that's
just me.<br></div></blockquote></div></div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto">Well said Ben. Your advice brought to mind this quote:</div><div dir="auto"><br></div><div dir="auto">"If a man will begin with certainties, he shall end with doubts, but if he will be content to begin with doubts he shall end in certainties."</div><div dir="auto">-- Francis Bacon</div></div><div dir="auto"><div dir="auto"><br></div><div dir="auto">Jason </div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div></div>
</blockquote></div></div></div>
_______________________________________________<br>
extropy-chat mailing list<br>
<a href="mailto:extropy-chat@lists.extropy.org" rel="noreferrer noreferrer" target="_blank">extropy-chat@lists.extropy.org</a><br>
<a href="http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat" rel="noreferrer noreferrer noreferrer" target="_blank">http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat</a><br>
</blockquote></div></div>
</blockquote></div></div></div>
</blockquote></div>