philosophy

LLMs dont think, they are statistical models that are good at prediciting what word might come next after another word

I admit that I don’t have a ton of formal experience with philosophy, but I took one course in college that will forever be etched into my brain. Essentially, my professor explained to us the concept of a neural network and how with more computing power, researchers hope to emulate the brain and establish a consciousness baseline with which to compare a human’s subjective experience.

Right now the ai capitalists are tricking you, thats ok, they have a lot of money to spend on propaganda, the current form of ai text generation is a highly advanced chatbot that is very effective at making us humans believe its close it consciousness. It isnt. We are a long way off consciousness in silicone, our technology just isnt there yet.

i think the through line from "sufficiently advanced computer + software" to "conscious brain akin to a human one" is basically made up nonsense conceived by SF writers and propagated as being actually real by tech bros. i don't think it's something worth taking seriously at all

As a data science undergrad, knowing generally how they work, LLMs are fundamentally not built in a way that could achieve a measure of consciousness.

Large language models are probability-centric models. They essentially look at a graph node of "given my one quintillion sentences and one quadrillion paragraphs on hand, which word is probably next given the current chain of output and the given input." This makes it really good at making something that is voiced coherently. However, this is not reasoning–this is parroting – it's a chain of dice rolls that's weighted to all writing ever to create something that reads like a good output against the words of the input.

The entire idea behind prompt engineering is that these models cannot achieve internal reasoning, and thus you have to trick it into speaking around itself in order to write out the lines of logic that it could reference in its own model.

I do not think AGI or whatever they're calling Star Trek-tier AI will arise out of LLMs and transformer models. I think it is fundamentally folly. I think what I see as fundamental elements of consciousness are just not covered at all by it (such as subjectivity) or are something I just find sorely lacking even despite the advances in development (such as cognition). Call me a cynic, I just truly think it's not going to come out of genAI (as we generally understand the technology behind it for the past couple years) and further research into it.

We're never going to develop artificial consciousness before the world burns up or humans nuke each other to death, don't worry

Pirate Severance

LLMs are dogshit tech and nothing to be worried about. capitalists will use them to replace jobs, but they won't be better at the jobs than humans, so in the long run they're going to come to nothing

Death to America

If you've ever watched Big Bang theory, you will begin to realize that much of the tech sector is literally run by people who think like Sheldon.

our pure arrogance wouldn’t stop us from creating another self and causing that self to suffer.

I think there's already some weak evidence supporting the notion that we'll do this with living neurons in a dish before we successfully simulate it in silico

Feeling like a certified Luddite these days

Big same, dawg

Human beings aren't all that special but biology is infinitely special. We have absolutely no fucking clue how consciousness works, how could we possibly hope to simulate it with a completely different mechanism?

Why does eveyone head to the trenches about consciousness this or that? Why even try to emulate the brain when transformers produce the results they do? All these systems need to do is convincingly approximate human behavior well enough and they can automate most professional jobs out there, and in doing so upend society as we know it. These systems don't need a soul to scab for labor.

And shit like neuralink exists and I think the next step will to be to offer that with a chatgpt integration or some dystopian shit.

Cannot wait to see a really shitty, stupid version of Upgrade play out in real time as lonely and depressed people start taking Grok advice from the honest to god voice in their head

You probably can simulate an intelligent creature in one of those 3ghz pentium processors. There are creatures with relatively simple neuronal systems that are capable of complex thinking, like the pack hunting velvet worms, or those tool using ants.

The thing is that the theory for how that works does not seem to exist. Chat got is a glorified Markov chain. It does not understand what it is saying. It cant it just calculates wich is the most likely token to produce next. It does not manipulate any categories or does anything resembling thought. Wolfram alpha does. That's the true state of the art. But there is a long way to go.

"Oh no my phone keyboard knows the next word I want to type almost like it's intelligent!"

Thee are not intelligent programs. They don't have memory of the past except for a few previous prompts. You are giving these tech bros way too mcu credit. It's like snake oil. Of course Doctor Health's snake oil can cure all ailments! Please buy it now, while it's this cheap! We just need ot get to the step where we tar and feather these fucks for their lies.

and our pure arrogance wouldn’t stop us from creating another self and causing that self to suffer.

People already have children all the time, this is just what humans do.

Hell, we collectively decided to slaughter en masse another collective group with feeling

My first thought was when we wiped out the Neanderthals

i don't know how they expect to 'establish a consciousness baseline' without a theory of information processing that can explain subjective experience. what is the algorithm that makes something experience? simulating a human brain on a computer will no more produce subjective experience than simulating a bladder will produce actual piss on your desk, as far as we know. it might tell us something about how consciousness works regardless, just as a bladder simulation can inform us about how a real bladder works even though you couldn't replace someone's actual organ with the simulation. regardless i am sure the information processing capabilities of humans will be more fully outclassed by computers eventually, or well enough to justify replacing paid human workers at least. although even current primitve LLMs require a lot of energy. like any other industrial revolution it will only be used to extract more profit instead of bettering society.

not that Id know from experience but dating on reddit is so much more fun when you get messaged and, unless they happen to use any of the key phrases you've learned are code for "i'm a fucking bot" you might spend 20-30 minutes "in conversation" before they start to plug their totally real only fans

Think about all the resources wasted on AI and then think about how many of these bots are running at all times just trying to grift some horny freak out of $3

how much AI is currently focused on AI-AI interactions as bots attempt to trick bots into subscribing to their onlyfans???

and our pure arrogance wouldn’t stop us from creating another self and causing that self to suffer.

People already have children.

our pure arrogance wouldn’t stop us from creating another self and causing that self to suffer.

I think there's already some weak evidence supporting the notion that we'll do this with living neurons in a dish before we successfully simulate it in silico

Feeling like a certified Luddite these days

Big same, dawg

If you've ever watched Big Bang theory, you will begin to realize that much of the tech sector is literally run by people who think like Sheldon.

i don't know how they expect to 'establish a consciousness baseline' without a theory of information processing that can explain subjective experience. what is the algorithm that makes something experience? simulating a human brain on a computer will no more produce subjective experience than simulating a bladder will produce actual piss on your desk, as far as we know. it might tell us something about how consciousness works regardless, just as a bladder simulation can inform us about how a real bladder works even though you couldn't replace someone's actual organ with the simulation. regardless i am sure the information processing capabilities of humans will be more fully outclassed by computers eventually, or well enough to justify replacing paid human workers at least. although even current primitve LLMs require a lot of energy. like any other industrial revolution it will only be used to extract more profit instead of bettering society.