The AC is incorrect. LLMs do it because their training data does it. Their training data includes lots of literature, scholarly humanities writing, journalism and ad copy. Naturally it sounds pretentious.

t is also why AI cannot replace any of the dangerous gross jobs people actually want replaced

https://renxhomes.ca/promise-r...

https://www.hitachicm.com/eu/e...

https://www.deere.ca/en/autono...

https://ntwist.com/blog/autono...

https://getsomatic.com/

https://blog.robozaps.com/b/hu...

It can, and is. Interaction with the real world is harder and takes longer because we don't have giant piles of training data sitting around ready to go like we do for images and text. There are also safety concerns and cost barriers that aren't there with things like text.

No post processing needed. LLMs sound like they do because that style was prominent in their training data and was scored highly by their trainers. If you ask most of the language models not to use excessive punctuation, lists, headings, polish your knob, and write at a grade six level, they generally sound reasonably human.

If you're watching some random thing on YouTube you should be wondering that regardless of whether it sounds like it was written by an LLM or not.

Photons don't interact with each other to produce composite particles.

Why? What would you call the minimum constituent parts of something? Atom (from Greek "indivisible") is taken. Part-icle (from Latin "piece" and "small") is pretty good.

There's an everything memory shortage. The reason, allegedly, is that OpenAI and then some of the other AI companies, bought up all the fab capacity to make mostly HBM. The memory fabs are fairly multipurpose: they can switch to make different things. The new ones probably can also but even if not, they will take pressure off the existing ones.

Anyway, you can type "new memory fabs under construction" into Google yourself and select whatever articles you think are relevant to your specific pecadillo.

I've got mine around somewhere too. It's in a beautiful aluminum case. It's also never connected to anything at 56k. 42k was the best it could ever manage on our phone lines.

Man, you guys are really butthurt that someone else might be doing something interesting hey?

1. You're being a silly pedant. They mean changing how advanced math is done. Mathematicians using AI to generate and check proofs, for example.

2. You might be surprised at how your pedantic interpretation might turn out to be true as well. It happened around the beginning of the 20th century when mathematicians engaged in a program to formalize the foundations of mathematics by adopting axiomatic set theory. It's happening now with a new formalization program advocating machine verifiable proofs. And there are ongoing things like the Langlands program that have changed the way we think about a lot of mathematics.

Just in case anyone reads the parent and thinks it might be true, it's bullshit.

Yes, it must be magic.

While I mostly agree, the one part I may disagree with is with documentation. It's not uncommon that documentation can be scarce for certain things which means that if someone can generate, review, and publish documentation for part of the project, I see that as a way of helping users that would otherwise have very little, just poor documentation, or in some cases no documentation at all.

Given that we live in a "market-driven" society, the correct conclusion is that companies are refusing to properly compensate skilled workers. As for the cause, the only rational conclusion is that there is insufficient competition in the energy distribution market. To encourage competition, the correct move would be to start breaking up large energy distribution companies like PG&E.

If only there was some way to process these flammable batteries in a way that their components could be reused again for batteries, like some sort of cycle that goes again and again, a "recycle" if you will.

I dunno, I'm just spitballing here.