This is a proxy for "the author didn't give a shit"..Not always, but often.

https://www.youtube.com/watch?v=OKnpPCQyUec

But let me tell you another possibility. What if there is no nature, but just our perception of it.

Yea, tree falling in the forest and all that...stated more explicitly..

In cases where LLMs are used to generate stuff, when push comes to shove, LLMs fails to deliver. This is fundamental.

For example, if you talk to an LLM regarding any subject, it will appear like a scholar initially. But if you push it deep enough, you will notice that it is just circling. This is same for any domain..

If this can only answer questions, then it fails this test. Because at least the question has to come from somewhere...

The right way to look at it is that LLMs help us to maximize the utility of the entirety of current human achievement/knowledge by discovering obscure connections in it.

No, no. Any random sentence generator can generate original idea. Actually it is said that randomness contain all the answers. You don't need a "meat brain" to do that.

What is the difference? Just curious...

Yes, this is the only way I use them. As a better search tool for the collective knowledge, and to do arbitrary queries on a complicated code base.

If there are any mechanical changes, like renaming something or such, I might use an LLM for it as well.

I think you cannot assume so, because pattern matching is not reasoning.

And the AI company for making the search program that searched through the data and found the solution.