Liste des Groupes | Revenir à theory |
On 10/13/2024 8:19 AM, Richard Damon wrote:On 10/13/24 8:53 AM, olcott wrote:I have only provided the source-code for DDD and the design of HHH.Although it is possible for LLM systems to lie:>
https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence)
>
ChatGPT does correctly apply truth preserving operations to
the premises that it was provided regarding the behavior of
DDD and HHH. *Try to find a mistake in its reasoning*
>
*The key premises are simply*
(a) the source code for DDD
WHich isn't the source code for the PROGRAM DDD
>
Try to run that program just by itself.
>
It won't work, you need to include HHH (and everything it calls) so that the "source code" for DDD needs to include the definition of all of that.
>
Sorry, you are just proving you don't understand what you are talking about.
>>>
void DDD()
{
HHH(DDD);
return;
}
>
(b) The design of HHH as a simulating termination analyzer.
Which gets the wrong answer.
>>>
https://chatgpt.com/share/6709e046-4794-8011-98b7-27066fb49f3e
Click on the above link to directly talk to ChatGPT about HHH
and DDD without logging in.
>
When you click on the link and try to explain how HHH must
be wrong when it reports that DDD does not terminate because
DDD does terminate it will explain your mistake to you.
>
You have taught Chat GPT this error as shown in this statement:
>
You have not shown how any details of exactly what I told ChatGPT
are incorrect.
Les messages affichés proviennent d'usenet.