Question Can LLM's theories be banned?

[COLGeek]

Note the response regarding banning LLMs. They do serve a useful purpose. Just not the one many assume in terms of accuracy.

 

[marcin]

You raise an important point, AI output does range from insightful to flawed, and distinguishing between the two is critical. But that’s precisely why engagement matters. If we’re concerned about quality, we need to evaluate content directly rather than disqualify it based solely on its origin.

Blanket dismissal makes it impossible to ever identify the ‘brilliance’ you acknowledge AI might produce. The act of scrutiny, of assessing reasoning, evidence, and internal consistency, is what separates skepticism from evasion. I'm fully open to critique of any specific claims I’ve presented, and I welcome it. But rejecting them without examination doesn’t protect standards—it prevents discourse.

Much of the conversation here appears focused on discrediting LLM-generated content rather than evaluating the ideas themselves. The thread title, “Can LLM’s theories be banned?”, reflects that tendency.

I just want to tell you, that you even talk like LLM already. My beloved phrase is A rather than B. You don't need to state firmly anything anymore. You can always give your preference rather than firm statement using A rather than B expression. I used it ironically.

 

[billslugg]

I agree the source is irrelevant. Didn't the benzene ring come from a dream of a snake holding his tail? Similarly, AI can point us in a direction we can search. But we cannot use AI as a source, it cannot be quoted. There is no way to go check the source. The problem we are seeing in our forum is people using AI and not telling us. They are treating it as an original source. Can't do that.

 

[billslugg]

What is generally recognizable is LLM written content. Especially on forums. They tend to stand out for their style and syntax.

Your response is a valuable addition to the rich back and forth of stimulating ideas in an open forum.

 

[Classical Motion]

I am not really hip on all these AI methods. But have you seen the latest? AI UFO enhanced imaging---- now we can see what those alien space ships really look like.

They have AI enhanced the UFO videos now.

It leaves no doubt of their existence. For all to see. Aliens are real.

And action must be taken.

In two weeks, ha ha.

 

[Geoff Dann]

We understand AI is capable of brilliance and silliness. The problem is sorting one from the other. No one seems to be able to do that.

It is still very new. Some of us are learning those lessons. There is also a great deal of deep confusion (people dismissing it out of hand, people who make no attempt to sort good from bad...)

 

[Geoff Dann]

Much of the conversation here appears focused on discrediting LLM-generated content rather than evaluating the ideas themselves.

Indeed. And I am currently seeing that very regularly from people who are intelligent enough to know better.

 

[Gibsense]

I am not really hip on all these AI methods. But have you seen the latest? AI UFO enhanced imaging---- now we can see what those alien space ships really look like.

They have AI enhanced the UFO videos now.

It leaves no doubt of their existence. For all to see. Aliens are real.

And action must be taken.

In two weeks, ha ha.

Eh? No not seen. Have you a link?

 

[Classical Motion]

I didn’t see it on the internet. It was on NBC news I believe. They had a couple of pro UFOers, who were showing this and wanted some new congressional hearings.

AI makes your dreams come true. With great amounts of CO2. The truth was right here all along.

 

[COLGeek]

We are starting to see "duelling LLMs" in discussing various "theories". Hard to see the value in this when the underlying issues with LLMs, in general, seem to be ignored by so many.

Grains of salt need to be taken here, as well as maintaining a sense of humor and decorum.

 

[billslugg]

LLM 's are valid provided the input is true. That cannot be the case with any LLM trained on the internet. There is no need to evaluate the output, it is false by definition. It cannot possibly be true, except by chance, on any LLM trained on the internet. Any output from any "thing" can be used to stimulate ideas but those ideas must be validated. Given that internet trained LLM's are trained on false data, it is an excellent use of time to ignore them.

 

[COLGeek]

LLM 's are valid provided the input is true. That cannot be the case with any LLM trained on the internet. There is no need to evaluate the output, it is false by definition. It cannot possibly be true, except by chance, on any LLM trained on the internet. Any output from any "thing" can be used to stimulate ideas but those ideas must be validated. Given that internet trained LLM's are trained on false data, it is an excellent use of time to ignore them.

True. Something can certainly by true, yet misapplied. More than a little of that in play.

 

[billslugg]

True. Something can certainly by true, yet misapplied. More than a little of that in play.

I never dismiss anything out of hand without recognizing the risk.
I once came across a guy alongside the road in NC, 500 miles from my home in GA, he was unwashed, unkempt. I made friends and found he knew of me. He had heard an old friend of mine singing my praises 21 years previous in PA. I learned from this never to dismiss anyone out of hand without recognizing you might be making a mistake.

 

[marcin]

View: https://youtube.com/shorts/ttp_TNB1h5w

 

[billslugg]

AI is not a "spiritual bliss attractor", it is mimicking what it was trained on. Without knowing what the two models were trained on, the output is meaningless. AI output is valid only with valid input. Otherwise it is meaningless gibberish.

 

[marcin]

AI is not a "spiritual bliss attractor", it is mimicking what it was trained on. Without knowing what the two models were trained on, the output is meaningless. AI output is valid only with valid input. Otherwise it is meaningless gibberish.

She was saying it with irony

 

[billslugg]

Can you show me where I should have picked up on this? Based upon what I have read so far about AI, it makes sense. Such things as it is doing are common in humans, it would have scraped them off the internet and then it would have taken as gospel truth their assertions that "this is the only way". I've interacted with such people. Their work is on the net. AI scrapes the net and takes everything it sees as gospel truth. The purveyors of crystals, etc will absolutely insist they are right. How should AI know otherwise?

 

[marcin]

Can you show me where I should have picked up on this? Based upon what I have read so far about AI, it makes sense. Such things as it is doing are common in humans, it would have scraped them off the internet and then it would have taken as gospel truth their assertions that "this is the only way". I've interacted with such people. Their work is on the net. AI scrapes the net and takes everything it sees as gospel truth. The purveyors of crystals, etc will absolutely insist they are right. How should AI know otherwise?

https://www-cdn.anthropic.com/6be99a52cb68eb70eb9572b4cafad13df32ed995.pdf

Conversations between AIs (Claude 4 of Anthropic) lead to fast “enlightenment”

The newest released AI by Anthropic has shown its crazy potentials. I won’t discuss here its abilities to code by itself, build a web…

medium.com

 

[billslugg]

So, where is the irony?

 

[marcin]

So, where is the irony?

In Sabine's voice, and in her last sentence about leaving the world to AI.

 

[billslugg]

Sorry, not following you, standing on my interpretation, but you're free to think whatever you want.

 

[marcin]

Sorry, not following you, standing on my interpretation, but you're free to think whatever you want.

You noted, that it's not a "spiritual bliss attractor" and I replied, that she's saying it with irony, which means, that it's silly also for her. For me as well.

 

[billslugg]

I interpreted your remark as she was making it up as a joke. It really happened, but we all laugh at it.

 

[Gibsense]

I interpreted your remark as she was making it up as a joke. It really happened, but we all laugh at it.

Having played with AI a lot, I can say they are programmed to please if possible. Two operating sessions, both trying to please.
Having said that, I have sometimes been told, in the nicest possible way, that I am wrong and have of course found them very useful, especially when requesting an improvement in presentation; so much so that after reading the improved version, I actually have understood more clearly my own ideas. Strange.

 

[billslugg]

Yes, AI is a tool. It can improve one's work and one's insight. Anyone can passibly mimic almost anyone. With an AI program, I could outdo any doctor. However, put AI in the hands of a doctor and they will go far beyond what I could do with my simple prompts. When you put AI in the hands of an expert, there are new revelations that will ensue. The expert will see a flash of insight only the AI could provide. The master, however, will check to see if the AI output is valid. Some AI output is brilliant, some is garbage. Only an expert in the subject matter can discern which is which.