> I don't understand why people are so fixated on the minds doing the mathematics being made out of meat.

I don't follow - are you surprised that mathematicians have social rules on how they interact with others?

You're definitely welcome to set up a journal that takes whatever types of papers you deem acceptable. It's not like they're preventing the dissemination of information by taking this stance.

Personally, I wouldn't hire a SW engineer who only showcases output from LLMs, and can't explain the code it wrote.

Would you hire a SW engineer that only showcases output from compilers, and can't explain the assembly that it wrote?

Since even the engineers that know what's going on aren't actually reading all of the AI output any more (or, if they are, they're not keeping up with their peer's output), why would you care? I don't think humans should waste time trying to understand their code, it's too slow and costly, and the understanding will be blown away the next time the AI changes it anyways.