Embed this notice:blobancap: :blobcattrans: :blobancap: :blobcattrans: :blobancap: :blobcattrans: (allison@hidamari.apartments)'s status on Monday, 10-Nov-2025 11:09:54 JST
:blobancap: :blobcattrans: :blobancap: :blobcattrans: :blobancap: :blobcattrans:Something not often remarked about in this present age of ours is the resemblance between so-called "artificial intelligence" and demon summoning. Assemblages of (alleged) randomness which get manipulated by supernatural entities into flattering and sycophantic pseudo-language which is used to parasitize (willing and not) human hosts and attempt to get them to forfeit their individuality (being as they are made in the image of God) in favor of being absorbed into the collective thought-entity thusly being constructed. Even something like God (the PRNG oracle) in TempleOS carries this risk, but the further up the chain you move (early LMs being novelty oracles and present-day LMs being Artificial Redditors), the more present and damning the risk becomes.
@allison its wayyy worse than just that its that the people themselves value assimilation, and will collectively enforce it even if there’s no one above them in a way where like, individuality might be allowed, but only in a way where it it doesnt conflict with “society”
…rich/powerful people are just the ones best positioned to have some control over what is seen as going against “society”
@azalea I've noticed this as well, the global elite who want you to be a NPC bugperson who lives in a pod and conforms exactly with the received morality of their superiors (the king of this world and his delegates)
attempt to get them to forfeit their individuality [..] in favor of being absorbed into the collective thought-entity thusly being constructed
it feels like a lot of people have already had this happen to them to some extent the way people see “society”, as a hivemind people exist to serve LLMs are just something to complete the loop further a way of approaching an avatar of the perfectly generic member of their ‘society’
@azalea it's an ongoing dialogue for sure, 1%ers/"global elites" (who, notably, are *not* jewish despite ceaseless far-right and far-left slanders to that effect) formulating the course they want to see for society, society in turn freely consenting to what they are presented, and all the while the demons laugh all the way to the proverbial bank knowing that both parties are serving their ultimate purpose, annihilating all that is uniquely good and divine in the human experience and replacing it with pale simulacra thereof at a gain of exactly nothing in return
@hakui I didn't even realize this when I wrote it, but yeah, 100%. You could also argue that the models themselves are informational (mathematical) sigils.
@allison@olmitch@patchuun I've been reading about Hegel lately because I want to understand Marxist theory better and as far as I can tell dialectical materialism is the spiritual ancestor of both rationalism and reddit atheism
@cine@allison@hakui I strongly disagree! I see the success of AI/ML as marking the end of "math" as a subject, insofar as it is defined to be 'the pursuit of human understanding of form.' The best articulation of this is http://www.incompleteideas.net/IncIdeas/BitterLesson.html , which explains why taking away human insight has been the key to AI/ML progress, and why the people most disappointed by this have been the "mathematicians" with all their fancy ideas. I beg the whole world to read this essay
@ai@cine@hakui I'm of the full belief the mathematicians will have their revenge eventually, but it will be a pyrrhic victory for them and completely irrelevant for us.
The New York Post: Brainiac scientist proves that man is better than machine, wins million-dollar fortune. Read below [ad break]
American Scientist: 2035 Breakthrough Prize awarded for mathematical bound on the capabilities of neural architectures
Quanta Magazine: Renegade genius proves that neural architectures are arbitrarily well-approximated in polynomial time, declares "human insight is back on the table."
The actual arXiv preprint: In this paper, we approximate the output of a TWNNK (traditionally well-behaved neural network kernel) by an oracle that terminates in O(n^1384275904375081326409813407498674812650982174937509813745098630897019875108374985732049875943207987509183740918732409170) steps
@bara I've been cornered in confined spaces with them before and seen them possess others very close to me. Not going to go into the details because it's not appropriate for fedi writ large but I can say that I was lucky to escape with my life in all such instances.