Athanase000

I think a realistic scenario is mass murders triggered by AI psychosis. Some weirdos start an AI cult and end up killing a bunch of randos because the AI said doing so would break the matrix or something similar. I am actually surprised this hasn't happened yet.

I think the person you are answering to was saying the same thing. They wrote "worth" instead of "worse".

I guess a weakness of the common benchmarks (artificial intelligence, gertlabs, and others) is that, from my understanding, they don't really cater to each language's specificity. If a language such as CL or Clojure boasts a superior REPL experience, that's entirely ignored by benchmarks, because they just look at the result of generic prompting. So maybe instead of writing countless blog posts on the superiority of such and such language, we should work on improving the benchmarking methodology. Then we would get more useful results.

I really don't understand this argument. The "opening tab" in Python has to be matched with an "absence of tab". I don't see any non-cosmetic difference between Python and curly brace languages.

Every time something is posted from Wolfram I realize I am the only guy who likes him. Yes, he does not state, or prove, theorems. He is studying alien mathematics. Nobody has any idea of what is going on, neither does Wolfram, but he is trying! In the world of weird computability experiments he is the one with the least worst understanding of the matter. I think the fact that we have all these computational stuff laying around and we have no idea what's going on, and probably never will (by virtue of the halting theorem or something similar), contrasted with the incredible amount of understanding that normal mathematics has brought forth, is mind-boggling. You can create a turing machine from very simple rules, and nobody on earth will be able to tell what it's doing. And most machines will be like that! Most cellular automaton will be like that! Computational irreducibility is a fascinating philosophical idea. Even if it is kinda fruitless. To me it looks like there is this infinite world of meaningless computation, and then somewhere you have this happy path of actual mathematics which is full of meaning and absolutely makes sense. And somehow humans stumbled upon this happy path.

For me the ultimate proof that AI will have become super-intelligent is not that it will prove results that mathematicians can already deal with, it's that it will be able to look at a Turing machine and tell you "Yep, that's how it behaves, it follows these rules, and will never halt because this and that". I realize this is impossible in general, but if it's possible to some extent, where it is way better than the people over at busy beaver challenge, personally my mind will be blown.

It does not matter to me. As long as the prose is good and the arguments are well-exposed I don't mind if an AI wrote it or a human. It would be different if we were talking about music, art or literature. But for science I care about the content, not who created it. The social aspect of mathematics was never alluring to me.

When I was in mathematics, reading other people's work was always more rewarding to me than doing my own. I would spend years on a question to get some progress, versus spending a few days or weeks reading a paper to enjoy something new. I understand not everyone is like me but I really welcome the omniscient mathematician machine people are predicting. I can't wait to ask all the questions I care about.

The way the LLM describes its own work at the end is hilarious.

Is there a site or github repository that aggregates all these non-trivial AI-solved results?

Have you used Claude Design? The site has all the hallmarks of it.

There is so many low quality projects that rely on vibe coded websites and vibe written copy, that this becomes a smell, to the detriment of projects like OP's which is genuinely interesting and looks like high quality, well-crafted software.