[ExI] Why stop at glutamate?

[Gadersd]

> But Gordon insists on GPT-4 has no understanding as if it was a mantra.

Mantras are indeed powerful. All one has to do if confronted with a contradiction is close the eyes and repeat the mantra until the problem dissolves.

GPT-4 can model human language. GPT-4 doesn’t understand words.
GPT-4 can solve math and coding problems. GPT-4 doesn’t understand logic.
GPT-4 can model spatial properties. GPT-4 doesn’t understand 3d space.

GPT-4 can explain all this. GPT-4 can’t think.

> On Apr 14, 2023, at 4:00 PM, Giovanni Santostasi via extropy-chat <extropy-chat at lists.extropy.org> wrote:
> 
> Or Gordon repeating GPT-4 has no meaning for the word apple when it came up by itself in saying we can use the indent of an apple as a way to distinguish an apple from a pear. Then it draws pears, and bowls (using 2 different ellipses one above the other to give the illusion of a rim), it knows where to apply the eyes of a person, and hair, it can draw little cute puppies. Or an AI and human exchanging a brain. All this without any visual training and relying on the internal representation of these objects. But Gordon insists on GPT-4 has no understanding as if it was a mantra. 
> Giovanni 
> 
> On Fri, Apr 14, 2023 at 11:54 AM Ben Zaiboc via extropy-chat <extropy-chat at lists.extropy.org <mailto:extropy-chat at lists.extropy.org>> wrote:
> 
> OK, we've established that Brent believes in 'physical properties' in 
> the brain, but can't explain what they are, and that Gordon believes in 
> 'referents' for words being physical objects, and can't explain how we 
> know what Smaug is.
> 
> We've also established that neither will budge from their position, and 
> constantly ignore any attempts to explain logically how those positions 
> must be false.
> 
> For all the talk of 'falsifiable hypotheses', there seems to be little 
> enthusiasm for abandoning them when they are actually falsified.
> 
> The only end to these arguments will come through exhaustion or boredom. 
> I see no possibility of an actual resolution.
> 
> Ben
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org <mailto:extropy-chat at lists.extropy.org>
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat <http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat>
> _______________________________________________
> 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/20230414/83b96dc6/attachment.htm>