I really want you to explain why you know better than the MIT researchers quoted in the summary who determined you are wrong. Most jobs can't be replaced by current LLM, that's what they found. Why do you disagree with it?

I said this:

You responded by saying rocks can think. Strong example of cognitive bias.

You're wrong here, and you refuse to cure your ignorance.

ok look. "Thinking" is defined with extreme mathematical rigor. It's not completely defined, but we can definitely rule some things out.

Of course, I don't have time to explain all the math here, so as an example I used the fact that rocks don't think. However, some people like you idiotically tried to argue that rocks do think. You are a moron! Why are you even arguing that??? Get off the internet for a while.

Read the book, your degree program failed you.

Your idea is that a NN which takes language as input, produces language as output... somehow has layers inbetween not related to language? Where do you come up with these ideas?

I stand corrected, this clearly shows that BYD cars are the best

If this is a topic you actually care about, there is a strong rec for this book, probably the best on the topic. You have the mental capacity to handle it.

Having "medicine security" is a good thing. You don't want all your medicine manufactured by a country that considers America an adversary.

It's a logical error to make an analogy, without also demonstrating that it applies to the situation. (Common mistake when people discuss legal things, which is why such conversations are often confusing).

No, I would not. I would care that you turn your brain on. If somehow you manage to do that, return to this comment, find the main point, and respond to that.

LOL Poe's Law applies to that comment. (In case you're serious, read some Karl Popper and help yourself. Or this one if you don't have the reading comprehension for Popper.)

How do you know? Do you have a citation for that?

Then why use the AI at all? Just present your own ideas.

Brevity is a gift.

"Using the outputs of LLMs to deliberately improve model quality" has nothing to do with model collapse. It reads like a PR release at best; at worst it's like LLM output.