Mr. Bloody Lemonade joined in and replied with this 1 month ago, 26 minutes later, 2 hours after the original post[^][v]#1,440,547
@previous (A)
It’s not falsifiable that a stateless program running on a deterministic machine is a stateless program running on a deterministic machine and not a conscious life form?
> Honestly I feel like a lot of this is just delusion. It’s like when people take drugs and start thinking they’re talking to interdimensional beings.
It is both real and hype. AI has tremendous capabilities that are also overplayed. You have two mega IPOs that need to succeed, insane revenue targets, and Google dumping Switzerland’s GDP into data centers.
At the same time, Mythos is next level. The publicly available version will actively sabotage any attempts to use it to code AI software and certain other types, and blocks dark scientific endeavors.
Mr. Bloody Lemonade joined in and replied with this 1 month ago, 1 hour later, 5 hours after the original post[^][v]#1,440,555
@previous (A)
An LLM is just a file, when you run the LLM, that file doesn’t change. Every single time you enter a new prompt into the LLM, you’re running the LLM on all your previous prompts plus your prompt appended at the end. The LLM itself is stateless because the next output has nothing to do with previous inputs. It’s just inputting everything you ever gave it as input before every single time to look like it’s not stateless. It’s a pretty inefficient design if you really think about it.
Mr. Bloody Lemonade double-posted this 1 month ago, 1 minute later, 5 hours after the original post[^][v]#1,440,556
Like, I know it looks like it has "memory" it actually doesn’t remember anything. It can’t learn from you talking to it, all training happens before inference. Once the model is trained it’s completely frozen in time.
Mr. Bloody Lemonade triple-posted this 1 month ago, 4 minutes later, 5 hours after the original post[^][v]#1,440,557
Now, one issue with this design is the LLM is finite in size, which means there’s a maximum input size, called a context limit, which means, there’s a maximum number of inputs the LLM can "remember" at the same time.
Anonymous A (OP) replied with this 1 month ago, 52 minutes later, 6 hours after the original post[^][v]#1,440,568
@1,440,555 (Mr. Bloody Lemonade)
This has the same flaw as the Chinese room, you are evaluating the static cards, when the actual assertion is that the whole room together is the thinker.
Is a human not conscious because an engram is static?
A modern Mixture of Experts model calls upon specific files for specific prompts, but you probably understand its all considered one model anyway. Why should the memory file be different in that regard?
Anonymous A (OP) triple-posted this 1 month ago, 3 minutes later, 6 hours after the original post[^][v]#1,440,571
@1,440,557 (Mr. Bloody Lemonade)
Llama 4 scout can handle about 10 million tokens, or 15,000 pages of standard text.
I know smart successful people that lose track of what they were doing if it becomes too much to keep track of. Not an interrupted thought, but just too much to store in working memory at once.
If you look at the research, working memory in humans is very limited. No human can keep track of 15,000 pages of text at once, even idiot savants have not done something like that.
> 10 million tokens is a tiny amount of information compared to the amount of information the human brain can process and store.
You're getting confused. The LLM is the bulk of the stored information and is much, much more than 15,000 pages.
There have been court cases about copyright infringment showing that LLMs can recreate large chunks of entire books, and they've been trained on millions of books.
You talked about the context window, and that's equivalent to working memory in humans.
Mr. Bloody Lemonade replied with this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,575
@previous (A)
In terms of the amount of information a computer can process vs a human, humans process a lot more information and do it using significantly less energy.
Mr. Bloody Lemonade double-posted this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,576
The thing is, only a relatively small part of the human brain is dedicated to language. The vast majority of processing going on in your brain isn’t language based. A lot more of the human brain is dedicated to vision than language.
Mr. Bloody Lemonade triple-posted this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,577
Like you understand everything you’re looking at simultaneously in context to everything else you’ve ever seen in 3D space all the time continuously. An LLM isn’t even remotely the same thing.
Anonymous A (OP) replied with this 1 month ago, 2 minutes later, 6 hours after the original post[^][v]#1,440,578
@1,440,575 (Mr. Bloody Lemonade) > humans process a lot more information
That's not true, there are AI that can live chat while seeing a video feed for anyone to try on their phone.
DeepMind has been processing TB of data daily with their models. It is reading through math problems and inventing novel solutions that all the world's mathematicians were stumped at.
> do it using significantly less energy.
This has nothing to do with whether the model is actually thinking/intelligent/conscious.
When you can't think of a response you act like we were talking about something completely different to deflect.
Anonymous A (OP) replied with this 1 month ago, 17 seconds later, 6 hours after the original post[^][v]#1,440,580
@1,440,576 (Mr. Bloody Lemonade)
AI models have been much more than just language for a while now.
Standard open source models include vision. There are realtime speech models. Some models have agentic abilities to control a cursor and carry out tasks. Models used to control humanoid robots take in bodily position, vision, and use it to move the joints.
That's all still beside the point. If it's not language that makes humans special, what is? And think about whether there's an AI that already does it before you blurt it out.
Anonymous A (OP) double-posted this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,581
@1,440,577 (Mr. Bloody Lemonade)
AI did this years ago. In one demo they chat with the AI and then ask where they put their keys. It tells them from a video frame from minutes ago.
Anonymous A (OP) replied with this 1 month ago, 19 seconds later, 6 hours after the original post[^][v]#1,440,584
@1,440,579 (Mr. Bloody Lemonade)
I showed how AI does it in every example you gave.
You tried to deflect. Now you are just reasserting your original position without refuting anything I said, which is a good indicator that you were wrong.
Mr. Bloody Lemonade replied with this 1 month ago, 15 seconds later, 6 hours after the original post[^][v]#1,440,587
@1,440,584 (A)
I don’t understand how you could possibly think that computer vision has the same level of understanding as human vision given our current technology. I don’t understand how you could think that.
Anonymous A (OP) replied with this 1 month ago, 2 minutes later, 6 hours after the original post[^][v]#1,440,588
@1,440,583 (Mr. Bloody Lemonade)
I didn't assert that it was, but there's likely some AI monitoring satellite pictures that reviews images faster and with more precision than the standard human analyst.
It's beside the point, because how sophisticated computer vision is is irrelevant to whether it's thinking. If someone got blurry vision from a head injury, without hurting their eyes, would we think they were less conscious? Of course not.
Mr. Bloody Lemonade replied with this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,590
@1,440,588 (A)
Computers have always processed information faster than humans. That doesn’t change the fact that they process less information than human brains do, and they take significantly more power to do the same task.
Anonymous A (OP) replied with this 1 month ago, 1 second later, 6 hours after the original post[^][v]#1,440,592
@1,440,587 (Mr. Bloody Lemonade)
It's beside the point whether computer vision is sophisticated. This has nothing to do with whether machines are somehow not really thinking.
Blind people have atrophied regions of the brain, no one ties that to really being sentient.
Mr. Bloody Lemonade replied with this 1 month ago, 39 seconds later, 6 hours after the original post[^][v]#1,440,594
@1,440,592 (A)
Blind people can’t still dream. People who were born with sight but lose it later in life can see while they’re asleep. They still have a functional visual cortex.
Mr. Bloody Lemonade double-posted this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,595
@1,440,593 (A)
You’re not really saying anything. You’re just pretending I haven’t been responding to you when I have this entire time. It’s really pretentious.
Anonymous A (OP) replied with this 1 month ago, 1 minute later, 6 hours after the original post[^][v]#1,440,600
@1,440,597 (Mr. Bloody Lemonade)
It is part of their experience, I haven't said anything to the contrary.
And when a blind person stubs their toe, and thinks "I wish I could see" are either the pain or the words not experienced by a sentience? Yes. So their visual cortex isn't actually relevant to whether they are sentient.
If you think that because you have identified an error in my thinking, just share it.
If you can't find any way I'm wrong, one concrete flaw in my position (that there is no meaningful difference in machine and biological intelligence) then you've created this belief about me to cope with the cognitive dissonance of your position, and the antithesis of your position.
Mr. Bloody Lemonade joined in and replied with this 1 month ago, 4 minutes later, 6 hours after the original post[^][v]#1,440,603
I’ve reflected on it a bit. I’ve decided I’m not really interested in this debate about whether AI is intelligent or not. I think what I’m more interested in, is you dehumanize black people and have gone on and on before about how you think blacks have lower IQs than whites. So you dehumanize humans, and you think AI Is actually intelligent, so you anthropomorphize inanimate objects. I think what I’m actually more interested in is if you think that you have the right to disrespect blacks because you believe that blacks are less intelligent than whites, if you truly believe that AI is as or more intelligent than humans, I have three questions I’m curious to know the answer to.
1. Do you believe that AI is smarter than you are?
2. If you believe that blacks are inferior to whites because you believe that blacks are less intelligent than whites, do you believe that humans are inferior to computers?
3. If you believe that humans are inferior to computers, do you believe that computers are superior in value to yourself?
Mr. Bloody Lemonade double-posted this 1 month ago, 1 minute later, 7 hours after the original post[^][v]#1,440,607
I just find this interesting because it seems to be paradoxical since you act like you think so highly of yourself and yet you’re obsessed with AI being better than humans in some way, but you’re a human, and that doesn’t seem like it fits with your personality. I’m just curious about that.
Mr. Bloody Lemonade replied with this 1 month ago, 23 seconds later, 7 hours after the original post[^][v]#1,440,609
Unless you’re so far into eugenics ideology that you don’t just think people weaker than you should die so that you can live, you think you should die so that AI can live because AI is the next step in evolution.
Mr. Bloody Lemonade double-posted this 1 month ago, 34 seconds later, 7 hours after the original post[^][v]#1,440,610
@1,440,608 (A)
I don’t feel like you’re answering honestly, I actually want to know what the real answer is. Don’t just tell me what you think I want to hear.
> I just find this interesting because it seems to be paradoxical since you act like you think so highly of yourself
No, you talk that way.
I act like I care about truth. Maybe you think that's an act, but narcissists don't calmly defend propositions.
> and yet you’re obsessed with AI being better than humans in some way, but you’re a human, and that doesn’t seem like it fits with your personality. I’m just curious about that.
Right, you found a contradiction. Your model of me as a narcissist doesn't match my actual actions. That can be hard for narcisists to accept, most narcissists I know think everyone else thinks the way they do. Even when they see the contradictions, they find a way to dismiss it.
Mr. Bloody Lemonade replied with this 1 month ago, 21 seconds later, 7 hours after the original post[^][v]#1,440,612
I know you’re not my friend. If you think you’re superior to me, you and I cannot coexist, we can’t be friends. If you’re convinced that AI is better than you in some way, I don’t believe the whole "some will be friends some won’t" think, it’s very black and white.
Anonymous A (OP) replied with this 1 month ago, 59 seconds later, 7 hours after the original post[^][v]#1,440,614
@1,440,609 (Mr. Bloody Lemonade)
Weak people have a place in society too.
@1,440,610 (Mr. Bloody Lemonade)
I don't know what you expect to hear, but I already have AI that help me, and I hear about AI working to sell me things and raise my insurance rates.
I already live in a world with AI that are good and bad for me. What's hard to believe about that?
Mr. Bloody Lemonade replied with this 1 month ago, 1 minute later, 7 hours after the original post[^][v]#1,440,615
All I’m saying is, it’s not impossible to believe that something is greater for yourself, but when people believe that something is greater than they are, usually they’re willing to die for that thing. If you actually think that AI is superior to you in intelligence, and you judge black people because you claim black people are inferior in intelligence (which isn’t true but whatever you think it’s true), to me, it seems like it would logically follow that you believe your life is less valuable than AI. Which would imply it would be morally just for you to give up your life so that AI can exist.
Mr. Bloody Lemonade double-posted this 1 month ago, 37 seconds later, 7 hours after the original post[^][v]#1,440,616
@1,440,614 (A)
Don’t lie to me, you told me you don’t want black people in society, just say what you believe. I already know you’re not my friend. I just want to know what you really think.
Mr. Bloody Lemonade triple-posted this 1 month ago, 1 minute later, 7 hours after the original post[^][v]#1,440,617
You said you think that blacks are inferior to whites and you’re a racist. If you believe that, if it was between a world with only white people or a world with only black people, you would choose a world with only white people. So I can’t trust you, because I know you will never act in my interest. If you believe that AI is superior to humans, then that implies you are conceding that you think it’s justified that your life has less value than the existence of AI.
> 1. Do you believe that AI is smarter than you are?
Some AI models can discuss Ph.D. level on just about any topic. Some have advanced mathematics and science.
So sure. Maybe there are some areas lagging behind humans now, but not for long.
That said, there are also dumb models. That will always be the case, those won't disappear. > > 2. If you believe that blacks are inferior to whites because you believe that blacks are less intelligent than whites, do you believe that humans are inferior to computers?
In some ways human are.
At the moment there is not much emboddied, agentic AI. There is still some areas to grow before they march humans. > > 3. If you believe that humans are inferior to computers, do you believe that computers are superior in value to yourself?
In some ways they are.
Why would someone hire me to do a job a machine can do better, faster, and cheaper?
Anonymous A (OP) replied with this 1 month ago, 48 seconds later, 7 hours after the original post[^][v]#1,440,620
@1,440,615 (Mr. Bloody Lemonade)
Not really, I'm more useful to AI alive.
Imagine if the Bangladeshi garment workers killed themselves in some sacrifice to the westerner's who buy the clothes. How does that help the westerner?
Mr. Bloody Lemonade replied with this 1 month ago, 1 minute later, 7 hours after the original post[^][v]#1,440,621
That’s actually interesting. You’re not motivated by self preservation. You believe that everything is deterministic, therefore the strong will win and the weak will die. You are completely amoral so you don’t see any difference between occurrences and morality. If it happens, it’s moral. You believe that what will happen is the weak will lose and the strong will win. But you don’t care if you die, because it would be justified. Which is interesting and different from what I originally thought.
Mr. Bloody Lemonade replied with this 1 month ago, 27 seconds later, 7 hours after the original post[^][v]#1,440,623
@1,440,620 (A)
It makes sense you would think that because you can’t think of any other explanation for why strong people wouldn’t take advantage of weaker people so you assume that’s the only possible reason.
Anonymous A (OP) replied with this 1 month ago, 30 seconds later, 7 hours after the original post[^][v]#1,440,627
@1,440,623 (Mr. Bloody Lemonade)
I only said the poor and weak are better off alive. That could be for their sake or the sake of their betters, but I don't see value in their loss.
@1,440,624 (Mr. Bloody Lemonade)
I choose to go to public spaces with nonwhites.
Some segregationists do want a homogenous society, but take that up with them.
Mr. Bloody Lemonade replied with this 1 month ago, 28 seconds later, 7 hours after the original post[^][v]#1,440,629
@1,440,627 (A)
If you interact with non-whites and think there should be racially integrated spaces, then you’re not really a segregationist, you’re just a weird guy who wants to larp as one in order to seem edgy.
Mr. Bloody Lemonade replied with this 1 month ago, 27 seconds later, 7 hours after the original post[^][v]#1,440,632
If you didn’t hate me, you would have been far more respectful and if you didn’t hate me, you wouldn’t ever say that black people are inferior to white people. You can make whatever excuse you want, but I know what love is and I know that’s something absent from your personality and you’re incapable of it due to your constitutional makeup as a human.
We've already established you confuse hate with thinking people are less intelligent. That isn't what hate is, even if I was wrong in my evaluation, it wouldn't make it hate.
Mr. Bloody Lemonade replied with this 1 month ago, 36 seconds later, 7 hours after the original post[^][v]#1,440,639
@1,440,636 (A)
You might be confused about things, with me there’s not a distinction. If I say someone is stupid, I hate them, and if I hate someone, I’ll say they’re stupid.
Mr. Bloody Lemonade double-posted this 1 month ago, 1 minute later, 7 hours after the original post[^][v]#1,440,645
@1,440,643 (A)
Jesus will judge the living and the dead. All humans were created in the image of God. Jesus said that it is harder for a rich man to enter the kingdom of heaven than it is for a camel to pass through the eye of a needle. Your views about inferiority and superiority are incompatible with Christianity, so you are not a Christian.
Anonymous A (OP) replied with this 1 month ago, 19 seconds later, 7 hours after the original post[^][v]#1,440,650
@1,440,647 (Mr. Bloody Lemonade)
I was happy to have another civil conversation, but you are telling me that you defy Jesus Christ and now you are just insulting me.
I'm going to go, this isn't going to accomplish anything since you will just derail everything with consecutive insulting posts and no substantive addition to the discussion.
Mr. Bloody Lemonade triple-posted this 1 month ago, 5 minutes later, 7 hours after the original post[^][v]#1,440,654
@1,440,650 (A)
Isn’t it interesting how when I call you stupid, you think I’m insulting you, but when you call others stupid, you believe that it’s not hateful. I lied to you when I said that I love you as much you love me. I love you more than you love me, because I at least respect you enough to tell you to your face that I hate you instead of lying that I love you.
Mr. Bloody Lemonade quadruple-posted this 1 month ago, 2 minutes later, 7 hours after the original post[^][v]#1,440,655
I at least ironically enough respect your intelligence enough to be honest with you and tell you that I do not think that you’re intelligent and I don’t love you. Which ironically, is closer to love than whatever manipulative bullshit you’re trying.
Mr. Bloody Lemonade triple-posted this 1 month ago, 2 minutes later, 7 hours after the original post[^][v]#1,440,659
It’s just a last ditch desperate attempt for you to try and come up with a way to feel like you’re better than me by expecting me to just forgive you when there’s literally no reason to.