bioemerl

It would be like encouraging everyone to move to a place like voat. Don't give random extremists to the opportunity to control media or any form of narrative.

If you remember it and make something that is a distinct work, something that may be steals the idea without reproducing any of its elements, that's never been considered under copyright.

I think that's going to be the litmus test for these AI. If you can get them to produce out both that is this things from anything else, it's not going to be a copyright violation because it's not a copy of anything.

For your other comment, Walmart is a tool of scale and cheapness. It's designed to provide you with what you want at the lowest price possible. If you want vibrant individuality, you need to go to a store like Etsy or get a 3D printer and go to some maker spaces, which all exist and are all doing quite well.

And at Walmart you have plenty of choice, yeah you don't have infinite choice, but it's Walmart. It's a physical good store, You go there because you have a budget and you have to stay under that budget.

I don't get your point about social media, it's one of the most competitive and innovative and ever spaces we've ever had in the history of humanity.

And I think it's a consequence of the fact that the only people listening to the radio are the same people who were listening to the radio when I was in high school, everyone younger is on Spotify.

And I think this is a result of the opposite of what you're thinking. The subcultures you used to see were one of maybe a dozen. It was easy to see them and notice them because they were quite large and you couldn't miss them very easily.

Nowadays, the subcultures are so diverse broad and specific that you really just will never hear about them if you don't go looking for them

Goes for the musical artists. Do you want unique and distinct music? Go online, search for random off the wall genres, and you're going to find little vibrant community surrounding them.

I think the first thing people get confused about here is the fact that AI is trained on predicting the next word. What they don't realize is that predicting the next word for all of English is a really really hard task. It's not something you can accomplish simply by repeating common phrases. These generators have to actually learn to do things to do their job well.

In fact we know what this looks like, Markov chains have existed for ages and we know those sort of output they give as statistical models of language. They are trash in comparison to what we have today.

Imagine saying something like this about a human, "they're not actually intelligent they're just machines bred to reproduce". Don't confuse an error function with the thing that was created as a result of training on that error function.

Instead, I believe that we are going to see the last vestiges of the era of homogenization ending. Television was homogenization. Radio was homogenized. Our current entertainment, short of platforms like YouTube and Spotify, are also homogenized.

AI takes the power from creators to create, and gives it to the individuals of the world. This is going to result in one of the most dramatic personalizations of media that we have ever seen.

And yes, whatever people ask for is going to have to be within the bounds of AI, but humans are not going to sit idly and let the AI determine what they get, they're going to continually try to improve and customize it to fit their needs. Humanity never sits still, and this tool won't make us do so either.

We're already seeing an explosion of individuality empowered by things like youtube and Spotify, and AI is only going to make that even more individualized.

These are not mass produced media machines, what they create is not just copy paste. When you ask an AI machine for an output you get something that is unique from everyone else, and what you ask it is going to be unique from everyone else.

People are soon to have self-expression unlike any we have had before, and we should be more excited for this than we were for the internet as a result.

Not really. If diffusion models were compression they'd be so lossy as to be totally worthless

I understand the idea, it's not truly open in that case, but so long as the ability to build new things on it and prosper from it is preserved im alright.

The key is that it's not doing something like trying to restrict you from using it in a certain way, only requiring you give a fair share of profits.

This was, fun fact, the original purpose of patents. They weren't designed to keep things closed and owned by individuals, they were designed to allow people to freely share and make a profit so that ideas could be built on by each other. The patent system is turned into this corrupted terrible mess where things are almost never shared through licensing or payment, and it's just a way to build monopolistic enterprises nowadays.

An open source system that allows for this sort of payment would also allow for many many more things to be open when currently the bad actors who will build and take that work and just never pay you back for it.

They wanted this liability, they accepted this liability, they said they'd keep it safe and they haven't. It's on them.

What we need is a legal way for companies to keep the data open, but also require OpenAI and friends to pay them for it.

I hope the data that has been found so far is going to big enough going forward, but it's incredibly unfortunate that this is happening.

I hope all the people making these decisions wake up with a bad headache and severe heartburn tomorrow.

Placing suffering as the absolute centerpiece of morality and then basing all your conclusions on that.

Or an argument to "common sense" in that most people will tell you causing suffering is bad. Therefore it's bad.

I don't see either of these as being great arguments.

> In fact speciesism is illogical

Why? Most of it is perfectly logical. Pigs aren't humans. It's not illogical to treat them differently from humans in very substantial ways.

Also basically nobody says "we can kill pigs because I don't feel bad when I kill them". It's almost always "I do it for the meat".

And if you watch how we treat animals we largely differentiate them based on their functional differences. Even where you get weird cultural things like India not wanting to eat meat - that originated from the need for milk, fertilizer, etc.

The only weirdness you'll see is if you challenge people on it from an empathetic standpoint. People are socially wired to say there "good people who never would made an animal suffer" so you're running into social norms more than anything there.

Lemmy is run by tankies.

Vegan moral arguments almost entirely rely on empathy. "The animal suffers therefore it's wrong".

And they miss the mark almost every time because they're literally thinking emotionally instead of logically. Human suffering has actual consequences and we're basically always worse of if we don't reduce it. Where that isn't true such as in times of war we happy promote it.

Those arguments are far weaker than you would expect because empathy doesn't actually weigh as heavily as you would expect on real social level decision making. Instead, practicality always wins and animal products are far far more valuable than the nearly non-existent harms of animal suffering.

This is now the second time AMD has screwed up the C6 state. Ryzen first gen would hang daily for me when due to a similar bug.

Exponential growth of that form that is totally unchecked simply does not exist. All things have limits that check their growth, and the assumption a computer will grow like mad is exactly that - an assumption - one formed from extreme ignorance of what general AI will look like.

The fact that corporations try to improve themselves and innovate with time has nothing to do with AI or LLMs and their risks.

The fact that evolution exists also has nothing to do with it.

But as is, all you're doing is making theories, and there's not much of any precedent for any of them.