Kind of speaks to this view.. https://www.anthropic.com/research/project-vend-2 <https://www.anthropic.com/research/project-vend-2>
From: Friam <[email protected]> on behalf of glen <[email protected]> Date: Tuesday, December 16, 2025 at 9:36 AM To: [email protected] <[email protected]> Subject: Re: [FRIAM] bigly I tend to think intelligence (or consciousness or sentience) comes from sensorimotor interactions. So the moral status of a gold fish (or even better an RNA protocell) is higher than that of super-intelligence ... depending, of course, on what we mean by "super". If it means "other dimensions" or even "meta", then *maybe* it has higher status. But if it means "higher order" or "further derived", then it has lower moral status than the *source* material from which it's derived. So I'd consider a creature with a high dimensional sensorimotor surface ... what? an octopus maybe? ... to have higher moral status than, say, a human with it's paltry 5± senses. Of course, this is fraught because if we decompose something like "smell" or "touch", it gets more complex than a single dimension, with a variety of different receptors for each "sense". Whatever, though. My intention is to call out the difference between many ways of interacting with the world versus sophisticated inference from a few ways of interacting with the world. So Claude/Clawd don't cut the mustard ... yet. What is the real difference between a rich type system with low throughput and a simple type system with high throughput? To boot, it's conceivable that creatures with intense/dense inferential organs are *prevented* from being in tune with the world. I did a bit of a deep dive the other day on the ~25-33% of people who experience significant adverse reactions to meditative and mindfulness practice (~6-14% long-lasting). Gazing at one's belly button isn't healthy for everyone. Of course, services like Claude and ChatGPT aren't merely LLMs. They have tools they can reach out to. But those tools are *also* derived, one might say of very (!) high order. I mean, Perplexity, Consensus, Asta, et al reach for sources relatively close to The World because they're classed as "scientific publications". But when Copilot reaches out to sites like humanreligions.info or religionunplugged.com, that's not so much interacting with The World as it's interacting with delusional fantasies, even if "agentic". [±] perhaps including things like proprio- or intero-. On 12/16/25 8:45 AM, Marcus Daniels wrote: > What the heck is "moral status"? If superintelligence doesn't get moral > status, then I think nothing will deserve moral status. > > -----Original Message----- > From: Friam <[email protected]> On Behalf Of glen > Sent: Tuesday, December 16, 2025 8:11 AM > To: [email protected] > Subject: [FRIAM] bigly > > https://www.404media.co/anthropic-exec-forces-ai-chatbot-on-gay-discord-community-members-flee/ > > <https://www.404media.co/anthropic-exec-forces-ai-chatbot-on-gay-discord-community-members-flee/> > > Jason Clinton, Anthropic’s Deputy Chief Information Security Officer (CISO) > https://www.linkedin.com/in/jason-d-clinton/ > <https://www.linkedin.com/in/jason-d-clinton/> > > Jason Clinton wrote 🙄: >> It’s quite a bit more complicated than you’d think: we don’t know what >> consciousness or sentience is, it’s called the hard problem of >> consciousness for a reason, [...] >> >> We have published research showing that the models have started >> growing neuron clusters that are highly similar to humans and that >> they experience something like anxiety and fear. The moral status >> might be something like the moral status of, say, a goldfish, but they >> do indeed have latent wants and desires, [...] > > I'm not posting to talk about how silly it is to ascribe consciousness to an > AI bot. Nor am I here to rant, again, about egocentric preemptive > registration. What irritates me about this is the word "highly". Maybe it's > just me. But for a decade or so, we've been pelted all day every day with > sh¡t Trump and his sycophants say. And either they *are* people with > impoverished vocabularies; or they know they're speaking to such. They tend > to intensify "weak" words like "similar". Because they're egoist, they *must* > intensify what they're saying. And because they don't have access to better > words, they have to intensify the few words they do have, e.g. "highly > similar". > > Were I Jason, I hope I could have had the self-awareness to eliminate > "highly" and simply write "similar". But a better option is to use stronger > words <https://www.cjr.org/language_corner/very-trump-tweet-intensifier.php > <https://www.cjr.org/language_corner/very-trump-tweet-intensifier.php>>. As > with everything about the TACO President, the mask he chooses tells us > something about what's under that mask. He pretends at strength because he's > inherently weak. > > Of course, maybe it's my pet peeve because I'm also guilty of it? .... No! > Stop it! "I got it, I got it! I know your damn words, alright?" > <https://youtu.be/zgvXtexdgAM?si=G5-3-epw235wuY7n > <https://youtu.be/zgvXtexdgAM?si=G5-3-epw235wuY7n>> > -- ¡sıɹƎ ןıɐH ⊥ ɐןןǝdoɹ ǝ uǝןƃ ὅτε oi μὲν ἄλλοι κύνες τοὺς ἐχϑροὺς δάκνουσιν, ἐγὰ δὲ τοὺς φίλους, ἵνα σώσω.
smime.p7s
Description: S/MIME cryptographic signature
.- .-.. .-.. / ..-. --- --- - . .-. ... / .- .-. . / .-- .-. --- -. --. / ... --- -- . / .- .-. . / ..- ... . ..-. ..- .-.. 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/
