[FRIAM] Free will—ghost in the machine or just clever wiring?

Mon Jun 9 15:33:23 EDT 2025

ChatGPT doesn’t have free will. Not even a little.

That’s because it’s deterministic—with only a sprinkle of pretend
randomness (the kind computers like to call “pseudo-random”). If you know
the random seed, you know exactly what “random” choice it will make. Every
time.

So, during training: give it the same data and the same random seeds, and
you’ll always get the same model with the same weights. No surprises.

And once the model’s trained, the output for any input is fully
determined—like a vending machine that gives you the same snack every time
you press A1, no matter how persuasive you are.

So, as much as it may sound like it’s making choices... it really isn’t. No
mystery. No free will. Just maths and memory.

On Mon, 9 Jun 2025 at 21:12, Marcus Daniels <marcus at snoutfarm.com> wrote:

> Could someone please take a definite position?   Can ChatGPT have free
> will or not.  If not, why not?
>
>
>
> *From:* Friam <friam-bounces at redfish.com> *On Behalf Of *Jochen Fromm
> *Sent:* Monday, June 9, 2025 12:01 PM
> *To:* The Friday Morning Applied Complexity Coffee Group <
> friam at redfish.com>
> *Subject:* Re: [FRIAM] Free will—ghost in the machine or just clever
> wiring?
>
>
>
> If you want to explain free will by entanglement then I would say free
> will is the opposite - a kind of un-entanglement or emergence.
>
>
>
> A biological system which grows while learning a language is an entangled
> system where two systems are merged into one, both entangled in the same
> structures. It is based on different codes stored in the same substance.
>
>
>
> Then you start to untangle them - for instance by self-consciousness - and
> get the biological animal on the one hand and the ghost in the machine on
> the other hand. A free will which is neither trapped by biological needs
> nor by advertising, brands and marketing would be the essence of a ghost in
> the machine, right? Although ghost buster Gilbert Ryle says such thing does
> not exist.
>
>
>
> -J.
>
>
>
>
>
> -------- Original message --------
>
> From: Marcus Daniels <marcus at snoutfarm.com>
>
> Date: 6/9/25 6:19 PM (GMT+01:00)
>
> To: The Friday Morning Applied Complexity Coffee Group <friam at redfish.com>
>
>
> Subject: Re: [FRIAM] Free will—ghost in the machine or just clever wiring?
>
>
>
> Here’s an idea that’s been helping me to procrastinate.
>
> 1. Suppose that spacetime is an embedding of entanglement.   An evolved
> quantum error correcting code (QEC) that enables a network to form
> geometries like the reality we see.
>
>
> 2.  Suppose the Big Bang the result of a unifying supermassive black
> hole.
>
>
>
> 3.  Like other black holes, it had high entropy.
>
>
>
> 4. That final black hole, lacking an exterior, launches a new universe.
>
>
>
> 5. The new universe might appear to be smooth in its geometric expansion,
> but that would only because of the embedded QEC.   It would be rich with
> unseen entanglement that was not subject to the QEC.
>
>
>
> 6.  In this view, universes could evolve or even be nested.   Universes
> with no or crude QECs would be unstable and prone to collapse.  Universes
> with strong QECs could have orderly environments where life could emerge,
> as Eric describes in his book.
>
> 7.  A Big Crunch would be like checkpointing a virtual machine.   The
> evolved QECs could still be in the checkpoint and cause the next version of
> the universe to inherit its desirable properties.  Maybe it would be like a
> junkyard with some interesting parts that would find novel uses in the next
> go.
>
> 8.  Speculating further, very sophisticated civilizations (after billions
> of years) might discover how to stack the deck to invent new metaphysics at
> the next Big Bang.   Simple beings, like humans – not being billions of
> years old -- might invent words for that like God.
>
>
>
> 9. The whole thing could be deterministic and not facilitate any free will!
>
> Now I should get back to work.
>
>
>
> *From: *Friam <friam-bounces at redfish.com> on behalf of Pieter Steenekamp <
> pieters at randcontrols.co.za>
> *Date: *Sunday, June 8, 2025 at 10:38 PM
> *To: *The Friday Morning Applied Complexity Coffee Group <
> friam at redfish.com>
> *Subject: *[FRIAM] Free will—ghost in the machine or just clever wiring?
>
> Seth Lloyd’s Turing test for free will (
> https://www.informationphilosopher.com/solutions/scientists/lloyd/Turing_Test.pdf)
> is to consciousness what EPR was to quantum physics: a challenge to the
> theory's completeness. EPR said quantum weirdness must hide something
> deeper; Bell said “let's test that”—and nature replied, “nope, it’s weird
> all the way down.” Nobel Prize, case closed.
>
> Lloyd asks: can we prove the mind is just machinery? His test says: build
> a machine that behaves indistinguishably from a human and believes it has
> free will. If you succeed—great. But failure proves nothing.
>
> Unlike Bell’s inequality, this test can only confirm, never deny. No
> ghost-busting here.
>
> Until then? It’s speculation. The Standard Model explains almost
> everything—except the quantum gremlins and how observation messes things
> up. So maybe the mind still has an ace up its sleeve. Or a soul. Or a bug
> in the code.
> .- .-.. .-.. / ..-. --- --- - . .-. ... / .- .-. . / .-- .-. --- -. --. /
> ... --- -- . / .- .-. . / ..- ... . ..-. ..- .-..
> FRIAM Applied Complexity Group listserv
> Fridays 9a-12p Friday St. Johns Cafe   /   Thursdays 9a-12p Zoom
> https://bit.ly/virtualfriam
> to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/
> archives:  5/2017 thru present
> https://redfish.com/pipermail/friam_redfish.com/
>   1/2003 thru 6/2021  http://friam.383.s1.nabble.com/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20250609/09e68a6e/attachment.html>