Sujet : Re: ChatGPT totally understands exactly how I refuted the conventional halting problem proof technique
De : news.dead.person.stones (at) *nospam* darjeeling.plus.com (Mike Terry)
Groupes : comp.theoryDate : 23. Jun 2025, 16:26:08
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <103brmh$1bfio$1@dont-email.me>
References : 1 2 3 4 5
User-Agent : Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101 Firefox/91.0 SeaMonkey/2.53.18.2
On 23/06/2025 16:01, olcott wrote:
On 6/23/2025 2:20 AM, Mikko wrote:
On 2025-06-22 14:42:13 +0000, olcott said:
>
On 6/22/2025 4:32 AM, Mikko wrote:
On 2025-06-21 22:48:01 +0000, olcott said:
>
int DD()
{
int Halt_Status = HHH(DD);
if (Halt_Status)
HERE: goto HERE;
return Halt_Status;
}
>
https://chatgpt.com/s/t_6857335b37a08191a077d57039fa4a76
ChatGPT agrees that I have correctly refuted every
halting problem proof technique that relies on the above
pattern.
>
If you like to play with ChatGPT you should ask it for
a solution to the halting problem.
>
I gave you the link showing exactly what ChatGPT agreed to.
https://chatgpt.com/s/t_6857335b37a08191a077d57039fa4a76
>
What an artificial idiot agrees to is not interesting.
A solution to the halting problem would be.
>
On 6/21/2025 6:21 PM, Mike Terry wrote:
> Is that the ChatGPT that does the peer reviewing for submissions
> for Proceedings of the London Mathematical Society?
>
> Seems like you're nearly there at last!
>
> Mike.
I have no idea why you quoted me on that, but just for the avoidance of doubt...
ChatGPT is a chatbot, and /obviously/ does not peer review submissions for any respectable journal - because it's a chatbot with no actual understanding of its subject matter, only recognition of patterns enabling it to produce plausible sounding responses to requests. I was probably being too subtle for you.
Mike.
ps. and of course, you're /not/ nearly there, even remotely.
You said:
Which version of ChatGPT are you?
ChatGPT said:
You're currently chatting with ChatGPT using the GPT-4-turbo model (often referred to as GPT-4o, the "omni" model). This is the most advanced model available as of June 2025 and is designed for high performance across text, vision, and audio tasks.
…