Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
AI-LLM Megathread
#1
Thread for AI and LLM-related snippets since this comes up a LOT and it's a topic I would comment on because of the stupidity of the educational sector about it.
Will repost snippets from Twitter here because this forum lets me do more without having to upload to the site itself. I hope to compile these into articles in the future.
Reply
#2
https://x.com/dearmadisonblue/status/205...8412304613

So the old "Hypercomputation" canard rears its ugly head. Let your boy Eugene explain this idea for you. It's bullshit that fails to understand what it means to "compute" anything. At its core is a stupid and wrong belief that language is in of itself "thought" or dictates thought by decree alone. So far as there is a legitimate notion of "hypercomputation" it would speak of what it would mean to compute things in some manner other than the algorithmic manner we have known. Algorithmic computing wasn't invented out of whole cloth. It's a family of problems that reach back to Antiquity and were known even before then, since there are records of algorithms written by the Egyptians and Babylonians. It was there that the true basis for our written record of mathematics came. Euclid was among those who compiled this into a digestible readable form that was elegant and established formally the study of mathematics.

Physicalism is not just an error. It's a mental illness that requires many assertions about reality that a child can disprove, and it demands everyone must obey this because "The Science" says so. It's insane.

What I do in my books, or at least allude to, is treat knowledge as a type of technology that can be judged on that basis, rather than this "master" faculty of humans that is given a ruling position it doesn't actually possess. That is, knowledge is this process that carries out an inquiry, which is something different from solving a problem which is what algorithms do. Humans inquire into the world not because they are programmed to follow a directive but because that is what we need to do to survive and what living animals did on some level to have a coherent cluster we could call thought at all. If this didn't happen, the animal would shamble around aimlessly. We can already see what the degraded condition of humans living under this curse has done. We simply give up and stop caring.

You don't get to use the verbiage of "solve" as if this hypercomputation is algorithmic while it is "above God". Such a thing would refer to a process that has nothing to do with computation or "solutions" in that sense. You can easily speak of knowledge that conducts this inquiry because we do this; and we do this again not because it is our purpose in life but because it has been a necessity. Comparing humans, who are lousy computers, to computers is missing the point. Computers do something well that we are very bad at, and computers are awful at this faculty of living that we did out of necessity. Computers would have a chronic problem with failure if they weren't kept as crude and simple as they could be; we engineer computers specifically so they don't exceed what the electricity is meant to do.

You would not need a hypercomputer to "solve the halting problem". All Turing's theory tells you is that some questions are not computable by algorithms. It says nothing about "fundamental knowledge" which we possess or can make claims about very easily. We can for instance claim, and offer proof of, knowledge that "all processes eventually halt" by some notion of what those processes are or anything that could lead to them. This is a crude statement and indefensible but we can surmise that for all of the processes that exist, we could predict their termination and the arrival of new processes. Nature and natural history have no concept of "transformation" as such; processes do not metamorph into other processes by some alchemy. We can speak of processes adopting new "code" from the surrounding universe and this is exactly what an actual computer must do; it must inhabit some machine in order to exist. All of the ways this can happen are predictable and are really considered part of the same process, if you can demonstrate that an algorithm is at work at all. That's the rub of course; the universe was never "computed" in that way, but a stupid and wrong theory of history insists it conforms to the crudest and lowest conceit of intelligence. That has always been the vanity of stupid people who should be ignored.

What is really missing here is that metaphysics became haram, and this has always been a choice. We are forbidden to ask questions about what things generally are, and this is the first step to abolish all comparisons and lock in a social order that decrees reality by diktat. That is not how the world ever worked, and this stupid and wrong theory causes needless suffering because it insists "this always works". If however you do have concepts of metaphysics, they are communicable in some language. A "hypercomputer" would really only be intelligible because the language can be rendered by someone or something into a meaningful thought in the genuine sense. That is, it wouldn't be doing anything magical. It would be a tool, just like a computer is only useful as a computer if its output is utilized by some entity that is not itself a computational process at some point. If computers were shuffling around the universe on autopilot, they would do nothing more than push piles of dirt around for no apparent reason so far as their activity is "thinking" anything.
Reply


[-]
Quick Reply
Message
Type your reply to this message here.

Image Verification
Please enter the text contained within the image into the text box below it. This process is used to prevent automated spam bots.
Image Verification
(case insensitive)

Forum Jump:


Users browsing this thread: 1 Guest(s)