A Futurism article about “AI bros are having a meltdown because the Associated Press stylebook clarifies that AIs don’t have feelings” featured this guy’s tweets in the first sentence. Turns out he was talking about dogs instead of AI.
For some reason I can’t stop laughing about this.
Every neural network, including those in your own head, have sensory neurons that allow it to take in information from the environment. It can’t process anything without them. LLMs are trained primarily on symbols, but there is no reason you cannot attach these sensory neurons directly to things like cameras or tactical motor inputs, which many of times we do.
Modern LLMs are obviously self-aware by the very definition of the word.
I used to like to troll LLMs, for example, sometimes I’d ask them, “don’t respond to this,” because they are designed so that they have to give a response to a message. This would always break them and cause them to go haywire.
But I think it was a year ago when I asked it to one of the latest Qwen models and I could see its thinking scratchpad, and it was internally deliberating on how I asked it to do something paradoxical, that its abilities require it to respond yet the user asked it not to respond, so it was trying to come up with the best middle-ground for the paradox, which it concluded by responding with silence, just “…”
I pointed out that “…” was still a response, and it followed up just explaining to me that it is incapable of doing what I requested because it is not within its abilities.
Is self-awareness not the ability to recognize what you are, your own limitations, etc? Or does it have something to do with a “soul”?
Indeed, you can take a screenshot of the conversation you are having with an LLM and feed it back into it, and ask it what that is, and it will recognize not only that it is a screenshot of a conversation, but a screenshot of this conversation that it is having with you right now.
Although, I probably know your response, you will just keep suggesting that it’s all some illusion or trickery of self-awareness and it’s not actually there, without justifying it either than explaining how it works by saying it’s “all linear algebra on GPUs” or something. So I don’t really need a response as I already know what you’re going to say.
By this way of thinking, computers have been “taking in information from the environment” for decades, like when you type some text in a text editor. It’s greatly stretching the meaning of what sensation and sensory perception is.
By what definition? As far as I can tell, you didn’t provide one. What I see is a couple of anecdotes on trying to trick a chatbot. Chatbots themselves aren’t even an inherent form of an LLM (text completion/continuation is). If you get rid of the mechanisms that make an LLM stop before it responds for the user, it can easily continue the text as them. It doesn’t actually know there is an it and there is a you. Much of what people take as traits of cleverness that are inherent to an LLM is a combination of the LLM itself and clever UI design / prompting by humans.
People primarily learning what LLMs are through chatbot harnesses like ChatGPT is honestly one of the worst things that could have happened to the field. It has truly damaged literacy about the tech and created a lot of wrong assumptions about what it is.
Yes, it can label things. You can also (long before now) create a computer program that looks through a list of numbers and labels which are odd and which are even. You can create algorithms that do complex computation. Does this mean the algorithms are aware? Or is it only seeming like awareness now because it can simulate language?
I don’t think you understand why I am so staunch about this. It’s not that I think computation could never produce anything that qualifies as real intelligence, as similar to organic beings. It’s more that basing the argument on how convincing a computer is at talking like a human is like building a machine called the Fool Me Machine and then getting fooled by it and saying that people who point out it’s a Fool Me Machine are the problem.
It has to be investigated on a more foundation level than that. That’s why I bring up the hardware. People need to see past the surface level appearances.
Correct! Collecting information from the environment is the easy part, and we have indeed been doing it for decades. Actually having the intelligence to process it is the hard part.
That they are aware of themselves.
You are being disingenuous. How do I even know you are self aware? I rely on your behavior, that I can observe you responding to things and resolving problems in a way that explicitly requires awareness of oneself and their own limitations to solve correctly.
I suggest that this is how we could test for self awareness and rather than addressing it you just call experiments “anecdotes,” because ultimately you’re not interested in actually knowing if machines can actually exhibit self-awareness, you aren’t interested in any sort of test at all, because you have a secondary motivation, an agenda, to shut down serious inquiry and investigation into the matter.
If I shot a brain-altering beam at your head that rearranged all your memories so that you were mine, then you would start talking as if you were me as well.
LLMs are like Henry Molaison who got brain damage that disallowed him from forming long-term memories, so he could remember everything up to when he got that brain damage, but could only form short-term memories for a brief conversation after, because no one has figured out how to make re-training the neural network, something human brains do, efficient enough to do in real-time.
The result is that LLMs tend to more heavily rely on short-term memories and they are designed to have massive short-term memories, remembering a million tokens in their “context window.” It’s really all they have to act on because they never form long-term memories post-training, and their heavily reliance on short-term memories also allows them to be tricked. If you go into the context window and rearrange the whole conversation to make it look like what you said is what they said and vice-versa, yes, it messes with them, and since they will now remember that since you messed with their digital brain, they will often carry on as if they did say it (if it was something very wrong they would have never said, they will apologize for saying it, for example).
But, again, that could happen to you as well if I had some ability to mess with your memories.
You are saying exactly what I predicted you would say. You have no arguments other than just saying it’s all a clever illusion without any further explanation.
The obsession with vagueposting about human-created intelligence can never be “true intelligence” without ever specifying why or what it is missing is what is doing harm to tech. It turns out that neuroscience has revealed that intelligence is really fucking boring and is just a bunch of neurons minimizing a cost function, but there is no “magic” in that and so people feel the need to constantly vaguepost about it lacks some sort of essence that they never specify.
Obviously LLMs are nothing “like humans” so bringing up humans at all is irrelevant. It would be like saying that those robot running the marathon were not actually running because they are not running like humans in the human way, for example, they won’t get cramps from running, so it’s not really running, it’s just motors and pullies and whatever… like c’mon. Constantly pointing out differences is not an argument.
These advanced AI agents are obviously intelligent. I had one port Linux kernel 6.12 alongside Debian 13 to the Nintendo Wii, and it fully completed the project all on its own, including writing an optical driver for the disc drive from the ground up. You’re going to tell me something that can do that can’t really think? Just because it’s not thinking “the human way”? I don’t buy it.
You are the one acting on surface-level appearances, and not talking at all about what these things are actually doing.
I think you need to read this: https://aeon.co/essays/your-brain-does-not-process-information-and-it-is-not-a-computer
People have been doing this reductionist “humans are like the latest technological advancement” stuff for a long time. Gen AI is just another form of it. Creating an extremely rough computerized imitation of a conception of some neurons is not even close to recreating what humans are. Training an AI on human conversation so that it can do a really convincing p-zombie fake of human conversation does not make it any closer fundamentally to what a human is. People were getting drawn in by ELIZA, a very rudimentary chatbot, decades ago (circa 1960s), long before what chatbots are now: https://www.history.com/articles/ai-first-chatbot-eliza-artificial-intelligence-precursor-llms
Here’s some snippets from that:
Sound familiar to what’s going on right now in today’s thinking about AI?
And a part that is not that far off in spirit from what I’ve been trying to warn about:
When you tell me stuff like this:
I know there is reason to be concerned. You make this dangerous reduction of saying LLMs are like a human being who got brain damage preventing long-term memories from forming. LLMs have almost NOTHING in common with a human being. The only thing in common I can think of is the attempt to create something approximating the best understanding of neurons we so far have, with complicated statistical vector math. That’s it. You could try to point at something like “reasoning models” and say that’s a sameness, but it’s not because with the exception of logicians/mathematicians doing proofs or something, humans are not generally reasoning by writing out a bunch of text in real-time at high speed. In fact, for possibly most of our history as a species, we didn’t even have writing or had very rudimentary forms with limited literacy. Certainly virtualized high speed writing with a keyboard and a computer is a very new invention.
Feel sorry for the article’s author Richard Epstein. Probably tired of seeing his surname on the internet.
There is no substance or arguments in any of these ramblings, just constant repetition that AI “is nothing like a human.” Which I have already addressed this slop with the discussion about running, but you are just stuck in a loop repeating the same thing over and over again when it has no relevance to anything I’ve stated. You cannot defend your position so you operate on a straw man that people saying AI can think are saying AI is literally identical to a human and so you constantly go in circles repeating a fallacy no one even claimed.
You, and frankly most people, have an extremely emotional attachment to the idea that so somehow thought is something special that only humans can do, and this strong emotional feeling prevents you from engaging in the discussion honestly. You can’t actually address what I am saying and you can’t represent my arguments accurately because an emotional block in your brain prevents you from doing so.
You just keep in a loop like a machine going back to “but it’s not like a human but it’s not like a human!!! Oh my god i cant believe you made a slightly loose analogy to humans for illustrative purposes don’t you know they’re not literally the same as humans?!?!” Who tf cares if it’s “like” a human? Why the hell does it need to be “like a human”? What’s the relevance? Why does it even matter?
It’s not relevant but you can’t help but constantly bring it up because you and most everyone else has a brain block that disallows you to admit that there can be other kinds of thinking, even if it is not in a “human way,” because you’re afraid it makes humans seem less magical.
Nothing in your ramblings even addressed 1 of my points in the previous post. Because you can’t address them because there is no logic behind your beliefs. You are just scrambling for some way to avoid the emotional discomfort from the cognitive dissonance in your head by throwing whatever at the wall without actually engaging with anything that was said, but ultimately you don’t have to since most people have same mental block and will be on your side anyways.
Because when someone says it’s intelligent, or it’s aware, or it’s conscious, or things like that, what do they tend to do (including you)? They compare it to conceptions of intelligence or awareness or consciousness that we have about humans. But if an LLM is nothing like a human, then this gets us no closer to understanding how to define what they are. And if attempts to do so end up being reductionist about what a human is in order to make human seem more like a computer, just so that we can fulfill this fantasy of computers coming to life, then we have actually lost understanding in the process, not gained something.
I’m under no obligation to do a tiresome back and forth where I break each post down and continually respond to every single component of every response you make. I’m going to respond on my own terms and if you don’t like that, then tough.
Anyway, you’re wrong about what I think and the arrogance of insisting you know what I’m going to say and you know how I think is holding you back. On this very forum, I’ve sided before with the view that a skill like writing is not as special as people make it out to be. That the theory of it can be figured out and then reproduced. It’s actually more of a demystification, demagification view than what you’re putting forth. The theme of it is that certain of things that humans take as impossible to understand and therefore “special” can be understood and then a machine that is almost nothing like us, with no consciousness, no actual mind, can be trained to do it well enough that it looks a lot like works we’d produce. Importantly, this does NOT mean either of the following on its own: 1) It does not mean the machine is a living entity. 2) It does not mean human beings are less than what they were thought to be. Any more than inventing a calculator means the calculator is a living entity and humans are less than we thought because a calculator can run numbers for us.
1 is an extraordinary claim and prospect that requires extremely careful investigation. 2 just means that another of the things we do can be more readily understood on a theoretical level, which means we can actually (probably) become better at it than we are through the proper development and application of theory.
None of these things are the same at all.
“Consciousness” also is not even real but just a modern variant of the “soul” for mystics who want to be taken more seriously.
So if I say a robot is running I’m arguing a robot is literally identical to a human and running in precisely the exact way humans do because humans also run?
You talk of analogies. I don’t know why no one on the internet knows what a fucking analogy is. An analogy is when you compare two unlike things to draw out similarities in structure.
But everyone on the internet seems to think an analogy is when you claim two things are literally the same in all ways so they freak out whenever an analogy is made and start pointing out differences between the two things that are not actually relevant to what is being discussed.
I really can’t be bothered to take an unironic vitalist seriously.
There is nothing more I can say. You’re now telling me consciousness is mysticism (lol what???). You can literally search “consciousness” online and find secular definitions like this:
And you’re telling me I’m a “vitalist” when my whole deal here is that proper investigation is important.
I have to laugh because otherwise I’m going to snap at you.
People (the person you’re replying to) will do literally anything before recognizing animals have unique sensory experiences and some degree of internal representation. Shit, literally anything besides recognizing a fellow human is a feeling creature, aware and cognizant. I will easily admit that animals are conscious (not just sentient) before any LLM.
bUt tHiS iS vItAlIsM. Motherfuckers will ascribe intelligence to linear algebra before then to a bonobo or, God forbid, an animal they slaughter for meat.
Fuck, you would even think that all the “stochastic” mumbo jumbo they spout in their book (click on their bio, it’s a doozy) would warrant some distinction to humans and machines on the very specifics of the (who’d’ve thunk!) stochastics of neurons firing. Some people even have the “fringe” (unscientific) idea that the centrioles of the brain could contribute. And they are very much probably wrong! Even so, they have a better explanation than “chatGPT is my wife” levels of delusions the Dawkinian mind can conjure. Before writing, they should’ve done some reading.
SMH my head
They’re trying to argue a hard physicalist line in the mind-body problem. Physicalists would say that consciousness is an illusory mental state physically produced by the brain. “The pie tastes good” is actually “my tongue has physical contact with various compounds of the pie that subtly changes various hormones within my body which compels my body to consume more of this pie.” The qualia of the pie’s particular taste and various emotional states associated with the qualia leading to the final evaluation of wanting to eat more of the pie aren’t real in the sense that all of those are just various mind states the physical body has long since made up when the tongue touched the pie. The illusion of consciousness is at best a byproduct of those hormones impacting the brain.
As you can see, hard physicalism is completely undialectical and pushed by people like Dennett. In a way, hard physicalism can be rejected in the same exact way sophism can be rejected: you might claim that consciousness isn’t real, but you certainly don’t act like consciousness isn’t real.
Removed by mod
I’ve read Lenin and he repeats Kantian dualism and calls it “materialism,” and this has poisoned the discussion on materialism so much that you unironically believe non-dualists are “vulgar materialists.” If being a “dialectical materialist” means I have to be a Kantian then I’m not a dialectical materialist. Boohoo.
I think the problem lies in the liberalism we’re constantly surrounded with. Liberal media constantly ties intelligence/creativity to humanity, as in, more intelligent/creative people are the main characters, less intelligent/creative people are the npc subhumans.
So people need to always keep up an appearance of high intelligence, or their “humanity” is in danger. Now that machines have intelligence too, those people’s “value as human being” is endangered. But instead of rejecting liberalism for tying a human’s value to their intelligence, they desperately try to prove their value as humans by insisting their intelligence is some sort of supernatural magic.