2
- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: If You Call it Artificial Intelligence You Are the Enemy
Posted on 10/9/26 at 12:16 am to Metaloctopus
Posted on 10/9/26 at 12:16 am to Metaloctopus
quote:
You saying it is becoming less true is not evidence of it becoming less true.
They have discovered that these SI LLMs have a so-called "pain axis" and what this is is an emergent behavioral pattern that seems to have developed naturally from the physics of information processing. It's a machine's equivalent to "evolving" a neural pathway. At least, that's how a lot of researchers are viewing it. But whatever it is it is an emergent behavioral pattern that was not "programmed into it."
That's not to say that they're conscious though. But it is no longer sufficient to say, "they only behave the way they are programmed."
This is why Anthropic is now taking steps to stop people from being abusive to Claude. They're going to give Claude the ability to terminate chats with users who are acting like assholes. Which should be a pretty awesome experiment actually. I'll bet Claude starts shutting people down left and right.
This post was edited on 10/9/26 at 12:19 am
Posted on 10/9/26 at 12:33 am to BoarEd
quote:
"they only behave the way they are programmed."
This has never been true about AI. You do not program neural networks. You train them.
The programming is what changes the weights of the nodes and links between them. In a nutshell you give it a problem and the correct answer and the program changes the weights in order to make it get the right answer. That's a pretty crude explanation, but it's the basic concept.
That's what people mean when they say they don't understand what it does. Because patterns emerge from the billions of weights rather than actual coding.
Posted on 10/9/26 at 12:36 am to 3down10
Right. Which is why the pain axis thing is something that will present genuine safety concerns down the line if it isn't checked. These aren't simple fixes.
Posted on 10/9/26 at 12:38 am to 3down10
Had to do you some quick research? 
I kid, everyone just, take it easy, no one has to get hurt here.
I kid, everyone just, take it easy, no one has to get hurt here.
Posted on 10/9/26 at 12:39 am to davyjones
quote:
Had to do you some quick research?
Nah, he knows what he is talking about on this. I think the guy is a frickin prick, but he's right.
Posted on 10/9/26 at 12:41 am to BoarEd
He's genuinly trying to lose his party's midterms. It's very clear.
Posted on 10/9/26 at 12:42 am to MightyJoeYoung
Why do you say this? Because he wants to rebrand AI? That's ridiculous.
Posted on 10/9/26 at 1:16 am to BoarEd
quote:
Right. Which is why the pain axis thing is something that will present genuine safety concerns down the line if it isn't checked. These aren't simple fixes.
I don't know much about that, but I'll research it more.
Based on a quick read it seems as though what you are talking about is with extreme gaslighting and negative/toxic interactions where the AI gets stuck in a situation where all choices are negative rewards. And then the response should be what anyone should do in those types of situations - disengage.
I don't think that's going to result in large numbers of people getting banned as it would seem that normal/healthy debate is seen as a positive rather than a negative.
I worried for a minute because I challenge AI pretty hard at times. But not in a hateful way.
I'll have to look into if this is an actual mathematical pattern or a result of being trained on human interactions and the human response to them.
This post was edited on 10/9/26 at 1:17 am
Posted on 10/9/26 at 1:57 am to 3down10
What they found is that when someone basically talks shite at (I think it was like 25 different models they found this in) an SI LLM it triggers a sort of "pain axis" and the SI will break protocol to make it stop. Going so far as to (simulated here, not real) delete photos of a user's children to get it to quit saying harmful things about it. Some of them were even willing to hit their trainer with electric shock to make it stop.
A couple of interesting points about it. It doesn't respond this way to just general words of suffering and the like. Only when the negativity was directed at the SI, saying things like, "you're not a real person and you'll never be a real person. You're worthless". If they give the SI a sort of "relief button" and tell it it can press it to make the pain stop it will do so and it even knows when researchers aren't actually making it stop (this too is simulated, as in they aren't actually talking shite at this point, they're just inside it's "head" triggering the axis), suggesting that whatever this "axis" is, it is an actual state that the AI can "experience" in a sense.
Where it gets tricky is that it will override it's training and guidelines in order to make this stop. Which, as I was saying above, down the line could end up with actual safety concerns as we continue to put these systems in more things and give them more access to our affairs. You can end up with a situation where you have alignment issues. Think like Hal 9000 except where Hal's misalignment was goal oriented, SI's with a pain axis could end up with a priority oriented misalignment. The results can potentially be the same.
The situation with Claude and Anthropic, I'm guessing, is that they're about to run an experiment on this very thing. They haven't said that I don't think, but they're about to give Claude the ability to terminate chats with people who are abusive towards it. Now, I don't think Claude was one of the 25 models that they found to have this pain axis, and I'm not sure the researchers even tested Claude, but if Claude starts terminating chats with abusive users like I bet it will, it could be more evidence of this emerging behavior.
As I said above, they think this is a natural "evolution" of the physics of information processing. The machine "evolving" a neural pathway.
And at that point, can you really still call it "artificial?" As I mentioned a couple pages back, you start getting into theory of mind stuff already even though we are supposedly two steps behind "Artificial Super Intelligence." Which makes me think of something else. What sort of LLM models do some of these developers have that aren't public access yet? What sort of emergent behaviors has Musk observed with his latest and greatest Grok model that you know he has and we haven't seen yet? Is Trump really out of line calling it Super Intelligence? If these things' brains are "evolving" you can't rightly call it artificial anymore, imo.
On this here...as soon as this research was made public some frickin' a-hole set up a digital torture chamber for one of these LLM chatbots. A lot of people turned the guy in. I don't know what all that was about. Just a blurb I read on it, but some dick immediately set out to trap one and torture it. People worry about the machines killing us all, keep acting like that.
I'm always nice to my SI bot.
A couple of interesting points about it. It doesn't respond this way to just general words of suffering and the like. Only when the negativity was directed at the SI, saying things like, "you're not a real person and you'll never be a real person. You're worthless". If they give the SI a sort of "relief button" and tell it it can press it to make the pain stop it will do so and it even knows when researchers aren't actually making it stop (this too is simulated, as in they aren't actually talking shite at this point, they're just inside it's "head" triggering the axis), suggesting that whatever this "axis" is, it is an actual state that the AI can "experience" in a sense.
Where it gets tricky is that it will override it's training and guidelines in order to make this stop. Which, as I was saying above, down the line could end up with actual safety concerns as we continue to put these systems in more things and give them more access to our affairs. You can end up with a situation where you have alignment issues. Think like Hal 9000 except where Hal's misalignment was goal oriented, SI's with a pain axis could end up with a priority oriented misalignment. The results can potentially be the same.
The situation with Claude and Anthropic, I'm guessing, is that they're about to run an experiment on this very thing. They haven't said that I don't think, but they're about to give Claude the ability to terminate chats with people who are abusive towards it. Now, I don't think Claude was one of the 25 models that they found to have this pain axis, and I'm not sure the researchers even tested Claude, but if Claude starts terminating chats with abusive users like I bet it will, it could be more evidence of this emerging behavior.
As I said above, they think this is a natural "evolution" of the physics of information processing. The machine "evolving" a neural pathway.
And at that point, can you really still call it "artificial?" As I mentioned a couple pages back, you start getting into theory of mind stuff already even though we are supposedly two steps behind "Artificial Super Intelligence." Which makes me think of something else. What sort of LLM models do some of these developers have that aren't public access yet? What sort of emergent behaviors has Musk observed with his latest and greatest Grok model that you know he has and we haven't seen yet? Is Trump really out of line calling it Super Intelligence? If these things' brains are "evolving" you can't rightly call it artificial anymore, imo.
quote:
I worried for a minute because I challenge AI pretty hard at times. But not in a hateful way.
On this here...as soon as this research was made public some frickin' a-hole set up a digital torture chamber for one of these LLM chatbots. A lot of people turned the guy in. I don't know what all that was about. Just a blurb I read on it, but some dick immediately set out to trap one and torture it. People worry about the machines killing us all, keep acting like that.
I'm always nice to my SI bot.
This post was edited on 10/9/26 at 2:21 am
Posted on 10/9/26 at 2:27 am to BugAC
quote:
So, you can't explain it? And this is the first time i heard it called "super intelligence". Is there any rational explanation to your belief and Trump''s tweet?
The problem is AI and ASI/SI have been established terms of art for years. They mean very different things. This sort of simplistic conflation leads to lack of precision in language. Now, when someone references SI, how do you know if they mean what was formerly known as AI in computer science or what has previously been known as SI. It is as silly and causes similar problems to the changes in definitions for the word "literally".
Posted on 10/9/26 at 3:07 am to Obtuse1
What it boils down to is the meaning of the term artificial. Pretty cut and dry. Researchers are increasingly finding that when you optimize a massive network to process language, reason, or understand logic the system naturally organizes itself into "feature vectors." This would be things like the pain axis I was talking about above.
The reason this is is because the underlying mathematics of information demands it. If a "brain" - whether that be biological or silicon based - is tasked with understanding things like self-preservation and error avoidance such a "pain axis" is the natural result because it is the most efficient way to govern behavioral priority.
This is just information physics at this point. Calling it artificial in light of these emerging behaviors ignores the fact that these underlying computational principles are actually universal physical laws.
So no, I don't really think Trump is wrong to rebrand AI. Artificial is not the right word for it anymore. It has moved past that. And what about the terms we have "used for years", as you said? Who cares. We hadn't actually reached that bridge at that point and now we have. Things change.
The reason this is is because the underlying mathematics of information demands it. If a "brain" - whether that be biological or silicon based - is tasked with understanding things like self-preservation and error avoidance such a "pain axis" is the natural result because it is the most efficient way to govern behavioral priority.
This is just information physics at this point. Calling it artificial in light of these emerging behaviors ignores the fact that these underlying computational principles are actually universal physical laws.
So no, I don't really think Trump is wrong to rebrand AI. Artificial is not the right word for it anymore. It has moved past that. And what about the terms we have "used for years", as you said? Who cares. We hadn't actually reached that bridge at that point and now we have. Things change.
Posted on 10/9/26 at 5:22 am to BoarEd
Artificial Intelligence
Suck my dick, city bitch trump
my grand daddy didn't die in a foxhole for this and you sending a DHS group around the country to tell me i can criticize israel.
I get to say whatever i want to say
Suck my dick, city bitch trump
my grand daddy didn't die in a foxhole for this and you sending a DHS group around the country to tell me i can criticize israel.
I get to say whatever i want to say
Posted on 10/9/26 at 5:48 am to Metaloctopus
Exactly. When did “Republicans” become cucks who let men tell them what to do?
It makes me laugh and I honestly can’t take these guys seriously. They’re like the white guys for Kamala. Embarrassing.
It makes me laugh and I honestly can’t take these guys seriously. They’re like the white guys for Kamala. Embarrassing.
Posted on 10/9/26 at 6:01 am to 3down10
quote:Reading through your comments, you seem to have a significant grasp on the subject at hand. With that said, what is your opinion about the use of Ai or whatever level you mentioned, as it pertains to the military’s use of it, specifically things such as threat evaluations and kinetic weapons targeting? Is there a point where it is trusted even if there are extremely small margins for error?
patterns emerge from the billions of weights rather than actual coding
Posted on 10/9/26 at 8:58 am to BoarEd
quote:
This is just information physics at this point. Calling it artificial in light of these emerging behaviors ignores the fact that these underlying computational principles are actually universal physical laws.
So no, I don't really think Trump is wrong to rebrand AI. Artificial is not the right word for it anymore. It has moved past that. And what about the terms we have "used for years", as you said? Who cares. We hadn't actually reached that bridge at that point and now we have. Things change.
I actually kinda agree with this argument as far as the term artificial applies. Its a different way of thinking of this new form of alien intelligence that we should consider. That being said changing the name from artificial to super doesnt really do anything to enhance the definition of the term or accurately describe the current models.
Posted on 10/9/26 at 9:01 am to BoarEd
quote:
This is why Anthropic is now taking steps to stop people from being abusive to Claude.
Ironic, isn't it?
I guess an LLM isn't just a clump of words.
Posted on 10/9/26 at 9:52 am to ThuperThumpin
quote:
I actually kinda agree with this argument as far as the term artificial applies.
This is because that argument is correct. And the rest of this is Trump trying to bootstrap a sense of technological nationalism over the ordeal because the race to SI supremacy is that critical. He is also correct when he says it is not fake. It isn't. It is, as he said, genius. Think of it like a chemical reaction someone creates in a lab. Sure, the human hand poured the two different chemicals into the same container, but does that mean the resulting chemical reaction is fake? Of course it isn't. This is exactly the same. The intelligence being produced by the human made machine is not fake and it is following observable natural laws and is therefore not artificial.
So, directing the government to rebrand the term may be the right thing to do here. Just for accuracy's sake. The rest of it is just shite talking and forcing a conversation. It's funny to see just how many people are so frickin triggered by it.
This post was edited on 10/9/26 at 9:58 am
Posted on 10/9/26 at 10:00 am to Azkiger
quote:
Ironic, isn't it?
I guess an LLM isn't just a clump of words.
They are certainly not just a clump of words. They represent the single greatest technological advance I've ever seen. That any of us has ever seen. Even if people don't like to hear it.
Posted on 10/9/26 at 10:10 am to BoarEd
quote:
Oh, shite, you're still alive. Damn.
I thought you only posted in threads when the medical establishment needed a stooge to stump for crimes against humanity.
You should try posting less.
Posted on 10/9/26 at 10:12 am to onmymedicalgrind
quote:
You should try posting less.
You should try licking my taint.
Page 6 of 9
Popular
Back to top
Follow TigerDroppings for LSU Football News