Yeah, I was talking about the Gemini blurb at the top of Google search results.

How many of those 900 million are being forced to use it?

Understand that the “AI” framing is pure marketing. It’s meant to get you to accept using LLMs to replace humans for every cognitive task. You have to decouple LLMs as a technology from the AI tech-bro hype train.

Oof. I think you need to find a better example than nuclear power.

LLMs existed before the AI craze. Sure, they’re useful in certain specific use cases. They aren’t a good substitute for a search engine or an artist or a therapist or any of the other myriad misuses now prevalent.

LLMs can make stuff resembling art. I’m not interested in it, though, because a human didn’t make it. What makes art art, what makes it interesting, is artists.

The room is filling with sewage. I’m not sure that this is a good thing.

It sucks as a search engine. When I’m searching for something, I want factual, correct results, not a hallucinated answer which is backed up by none of its sources.

When I click through search results, I’m reading, evaluating the sources, and synthesizing the information into something which I am reasonably satisfied is true enough. I don’t feel comfortable handing that cognitive task off to a machine. The machine’s job is to help me find the sources, not tell me what the truth is.

Having a machine that tells you what is true is the most dystopian shit ever, and it would’ve sounded completely insane only 5 years ago.

I think that’s just intelligence.

Ugh, yes. I try to always remember to update with what I did to fix it. Although if I’m honest, what probably happens 95% of the time is I just forget about the post entirely.

Money is no object to AI companies. That’s one of the things that alerted the booksellers, in fact—masses of books were being purchased without regard to price or subject matter.

Yeah, model collapse is worrying, but I don’t see how they can do anything about it due to how LLMs work. There have been some articles written about this drawing an analogy to the grey goo problem in nanotechnology. The end result is that the free web becomes essentially useless and knowledge becomes balkanized and paywalled.

The answer is yes in all cases. Well, almost all. They don’t want books printed after 2022, because those might have AI-generated text in them. Anything else is fair game.

Are you skeptical that the books being destroyed are actually rare? Or are you skeptical of the existence of rare books?

“Rare” means “difficult to acquire”. Under your definition, the only books that would qualify would be ones in museums.

I’m only interested if I get AOC feet pics.

Especially since the articles make it clear that the AI companies don’t care about rarity or quality. It’s just a dragnet to get as many books as possible. Sure, a lot of the books are going to be garbage, but some of them are going to be valuable, too.

The AI must be fed, and the only criterion is the book has to be published before 2022.

I do most of my shopping through eBay, although some of the sellers use Amazon for fulfillment. They’re hard to avoid entirely.

A digitized copy which isn’t shared with anybody else? Yes, I would disagree.