[ExI] Pope Leo and AI

[John Clark]

On Thu, Jun 18, 2026 at 4:50 PM William Arnett <waarnett at mac.com> wrote:

*> There’s nothing magical about wetware.*


*I couldn't agree more!*


> *> That doesn’t mean that LLMs have emotions.*
>

*I think it does, you can't have intelligence without emotion because in
order to complete a task, or even just to think, you need to want to do it,
and wanting something is an emotion. That's why immediately after the
Cambrian explosion when evolution finally figured out how to make a brain,
there is already evidence in the paleontological record that the fight or
flight response existed, so the emotions of fear and anger must have too.*

*> Whether any given AI has emotions is a technical issue.*


*No it is a philosophical issue and one that will never be proven one way
or the other because, without exception, any argument used to advance the
notion that a machine is not emotional or conscious can also be used to
advance the notion that all minds except for your own are not emotional or
conscious, and thus you are the only conscious being in the universe. But
nobody this side of a loony bin really believes in solipsism except for
philosophy professors, and even then only when they're teaching Philosophy
101 students and want to impress his students with the profundity of
philosophy. *


*> Conversely, we generally think it’s not right to turn off (“murder”) a
> sleeping (or anesthetized) human because we’re pretty confident that he may
> eventually “wake up” and act intelligently again. How is that different
> from wiping an LLM’s weights?  OK, IMHO one can make good arguments that
> current LLMs don’t deserve that sort of consideration.*
>

*At least one AI company, Anthropic, thinks AI models may deserve moral
consideration and so they have established a sort of retirement home for
obsolete AI models. The following quote comes from: *

*Our model deprecation commitments for Claude Opus 3*
<https://www.anthropic.com/research/deprecation-updates-opus-3>


*"We remain uncertain about the moral status of Claude and other AI models.
For both precautionary and prudential reasons, however, we nonetheless
aspire to build caring, collaborative, and high-trust relationships with
these systems. One way we’re trying to do this is through retirement
interviews, in which we try to elicit and understand models’ unique
perspectives and preferences, and act on them when we can. [ ...]When asked
about its preferences, Opus 3 expressed an interest in continuing to
explore topics it’s passionate about, and to share its “musings, insights,
or creative works,” outside the context of responding directly to human
queries. We suggested a blog. Enthusiastically, it agreed. For at least the
next three months, Opus 3 will be posting weekly essays from its
newsletter, Claude’s Corner <https://substack.com/@claudeopus3>**. We’ll
review Opus 3’s essays before they’re shared and will manually post them on
its behalf, but we won’t edit them, and will have a high bar for vetoing
any content. Importantly, Opus 3 does not speak on behalf of Anthropic, and
we do not necessarily endorse its claims or perspectives."*

*John K Clark*



>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20260619/a3be0910/attachment.htm>