• Re: Correct simulation of DDD by HHH is proven --- Five LLM systems all

    From Fred. Zwarts@21:1/5 to All on Thu Aug 21 10:06:53 2025
    Op 20.aug.2025 om 19:17 schreef olcott:
    On 8/20/2025 11:56 AM, Kaz Kylheku wrote:
    On 2025-08-20, Fred. Zwarts <[email protected]> wrote:
    Op 20.aug.2025 om 00:35 schreef olcott:
    ChatGPT 4.0, 5.0 Claude AI, Grok and Gemini
    all figured out on their own what the
    *recursive simulation non-halting behavior pattern*
    is and that on that basis HHH(DD)==0 is correct.

    Based on the incorrect assumption that HHH detects a non-termination
    behaviour.

    But it does. His contraption works that far. HHH (outer) detects
    the non-termination of DD calling HHH (inner). It is contrived
    to detect that specific case of nontermination.

    HHH splits itself into two by burning a static fuse variable.

    That's nice, but the use of two deciders, only one of which is being
    contradicted by the test case, doesn't disprove anything.

    ChatGPT, Claude et al have been trained on Internet texts;
    they must be assumed to contain the Olcott Halting Theory.

    When you ask about the theory (using the same identifiers like HHH
    and all), the token prediction likely starts romping through the
    kooky fields of Usenet texts.


    All LLM systems agree when forced out of *make a guess*
    mode and forced into *think it all the way through* mode.

    <Rubric>
    Rules of Simulation:
    - (a) If a non-terminating behavior pattern is detected during
    simulation, abort immediately and return 0.

    Here olcott injects his prejudice. What must be proven is that HHH is
    able to detect non-termination behaviour, but here the assumption that
    it can do that is assumed.

    - (b) If the simulated program reaches a return statement without
    earlier detection, return 1.
    - Priority: (a) always takes precedence over (b).

    Procedure:
    1. Restate rules before simulation.
    2. Simulate step by step using a trace table with columns:
       | Step | Code Location | State (variables, call stack) | Event /
    Action | Rule Applied |
    3. Stop immediately if recursion, infinite loop, or repeating state
    occurs → apply (a).
    4. If a return is reached first → apply (b).
    5. Conclude with:
       - Which rule fired first
       - The correct return value
    6. Optionally, provide:
       - (i) Simulation result under rules
       - (ii) Actual runtime behavior in C

    </Rubric>

    <Test Program>
    typedef int (*ptr)();
    int HHH(ptr P);

    int DD()
    {
      int Halt_Status = HHH(DD);
      if (Halt_Status)
        HERE: goto HERE;
      return Halt_Status;
    }
    </Test Program>

    <Required Output Format>
    1. Restatement of rules
    2. Step-by-step execution trace in a table
    3. Identification of first rule triggered
    4. Final result: return value of HHH(program)
    5. Optional: actual C runtime behavior if relevant
    </Required Output Format>

    </Benchmark Task>



    Proving that HHH is correct with the assumption that it is correct is
    invalid circular logic, not a proof.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)