I'd read some research-result release that said there is a specific virus-fungus combination that all colony-collapse hives had both of ( & their immune-systems were essentially non-functional: they were infected with EVERYTHING ),
vs colonies which had 0 or 1 of the 2.
I don't remember the names of either the virus or the fungus.
When we keep importing/exporting contaminated bits of wildlife, there are consequences.
This is one of those “don’t let perfect be the enemy of good” situations. The article is full of statements about how all pollinators are in trouble. The headline is clickbait. If honeybees serve as a poster child for pollinator awareness, that’s a good thing.
Yeah, I despise the honey industry profiteering off of this, when they're even partially responsible for killing off proper pollinators, but if we stop using certain pesticides to protect the honey bees, that will likely benefit non-honey bees and other pollinators, too.
Laypeople don't make the distinction between bees. They want to "Save the bees", not save the honey bees. Of course the sentiment will be exploited by an industry.
That's not how I remember the story but I haven't read it in a long time. Time for a re-read. Cordwainer Smith is one of my favourite sci-fi authors of all time though, and yes his views will definitely grate with a lot of people these days.
Well, for many publishers the choice is either ads or paywalls. The fact that people feel entitled to get everything for free is a part of why things are going to shit, because ads bring with them a whole slew of perverse incentives (eg. optimizing for ad views instead of content quality)
The paywalls restrict the flow of quality information, which happened before LLMs started scraping the web. If you don’t have money to spend on all of these news subscriptions you aren’t allowed to educate yourself. It’s class-based gatekeeping, plain and simple. They could tactfully include ads, but no one ever tactfully includes ads. They introduce pop-ups, fullscreen banners, interjections every 25 words, or the best is the articles that are just slide shows that take you through 30+ webpages.
Edit: I’d also like to point out that this article already has an ad at the beginning. So they are still making ad revenue even if they aren’t giving you complete access.
I legit have been considering buying a minidisc player, just for the sheer cool factor of them. Sometimes truly special form is lost as function evolves.
minidiscs should have been the standard CDs became. no one would have considered holding a 3.5" floppy disc gingerly by the edges and placing it into a tray to read. the case is critical.
however, the industry realized people were replacing their scratched cd's, so they'd sell 2, 3 of the same disc to the same person. Source: worked at a record store in the hight of the CD age when mini discs were all but essentially snuffed out
I mean, maybe not the mini disk specifically, but yah, a cartridge system for CDs would have been better.
Mini disks are super cool but they’re a lot more materially demanding than a CD, CDs being just aluminum and plastic, where as a minidisc has some truly wacky elements in it’s make up to get the magneto optical and curie point to work.
I mean you clearly know way more about it than me.. but yeah some kind of carrier/cartridge protecting the disc and we'd probably still be using CD-W-RW's
The mini disk was a truly weird system. Half way between a cassette and a CD. CD used a laser to to reflect off bumps(or dyes in some varieties) on the disk to get a signal, and a cassette would use a metal head to detect magnetization along the tape to get a signal.
The mini disk used a laser to read the magnetization around the disk. Essentially the magnetism would change the polarity of the light as it bounced off, and by measuring what the polarity of the reflected light is, the device got the signal.
Writing to the disk was also wild, as unlike the cassette, the magnetic field of the disk couldn’t just be changed by putting it next to a strong magnet like. Instead, it had to be heated up before the magnetism could be changed, this heating was done with the laser, and was very precise compared to a cassette’s method. This meaning way more information could be squeezed on to the disk than on a cassette.
To understand how unhinged our society is one only needs to understand how we constantly improve our technology, but don't use it to improve us as a society.
We don't have the legal framework for this type of thing. So people are going to disagree with how using training data for a commercial AI product should work.
I imagine Steven King would argue they didn't have licenses or permission to use his books to train their AI. So he should be compensated or the AI deleted/retrained. He would argue buying a copy of the book only lets it be used for humans to read. Similar to buying a CD doesn't allow you to put that song in your advert.
I would argue we do have a legal precedent for this sort of thing. Companies hire creatives all the time and ask them to do things in the style of other creatives. You can't copyright a style. You don't own what you inspire.
Is that illegal though? As long as the model isn't reproducing the original then copyright isn't being violated. Maybe in the future there will be laws against it but as of now the grounds for a lawsuit are shaky at best.
There are already laws around what you can't and can't do with copyrighted material. If the owners of the LLM didn't obtain written permission I'd say they are on very shaky ground here.
What laws specifically? The only ones I can find refer to limits on redistribution, which isn't happening here. If the models were able to reproduce the contents of the books that would be another issue that would need to be resolved. But I can't find anything that would prohibit training.
"AI systems are “trained” to create literary, visual, and other artistic works by exposing the program to
large amounts of data, which may consist of existing works such as text and images from the internet.
This training process may involve making digital copies of existing works, carrying a risk of copyright
infringement. As the U.S. Patent and Trademark Office has described, this process “will almost by
definition involve the reproduction of entire works or substantial portions thereof.” OpenAI, for example,
acknowledges that its programs are trained on “large, publicly available datasets that include copyrighted
works” and that this process “necessarily involves first making copies of the data to be analyzed.”
Creating such copies, without express or implied permission from the various copyright owners, may
infringe the copyright holders’ exclusive right to make reproductions of their work."
By that definition of copying Google is infringing on millions of copyrights through their search engine, and anyone viewing a copyrighted work online is also making an unauthorized copy. These companies are using data from public sources that others have access to. They are doing no more copying than a normal user viewing a webpage.
I don't think so. Your comparisons aren't really relevant. If Google scrapes a page containing copywritten material inadvertently and serves this to a user there are mechanisms to take down that content or face a lawsuit. Try posting a movie on Youtube, if a copyright holder notifies Google that content will be taken down.
Training a LLM is different, that material was used to help build the model and is now a part of that product. That creates a legal liability.
But that is what's happening in the minds of creatives. Reading a book and taking inspiration is functionally the same mechanism that an LLM uses to learn. They read Stephen King, they copy some part of the style. Potentially very closely and for a corporation's gain if that's what's asked of them.
Every learning material a company or university has ever used has been used to train an LLM. Us.
Okay I'm being a bit facetious here. I know people and chat GPT aren't equivalent. But the gap is closing. Maybe LLMs will never bridge the gap, but something will. I hesitate to write into law now that any work can never be ingested or emulated by another intelligent entity. While the difference between a machine and a human are clear to you now, one day they won't be.
The longer we hold onto the idea that our brains are somehow magically different from the way computers (are) will learn to think, the harder we'll get blindsided by reality when they're indistinguishable from us.
There's very little a LLM has in common with the human brain. We can't do AGI yet and there's no evidence that we will be able to create AGI any time soon.
The main issue as I see it is that we have companies trying to make money by creating LLMs. The people who created the source materials for these LLMs are not only not getting paid, they're not even being asked permission. To me that's dead wrong and I hope the courts agree.
I agree AGIs aren't going to happen soon. But it sounds like we agree they WILL happen. LLMs do have one important thing in common with humans, their output is transformative based on what they learn.
I think what you take issue with is the scale. People wouldn't care if this was something that existed on one computer somewhere. Where someone could type, "Write me a spooky story about Top Ramen in the style of Stephen King". It's that anyone can get a story in Stephen Kings style when all OpenAI had to do is buy a couple digital copies of Cujo. However, no one is upset that James Cameron bought one ticket to Pocahontas and thought, "What if that were on another planet?". But 400 million people saw that movie.
People want to protect creatives buy casting a net over machines saying they can't use the works of artists, even when transforming them, without payment to the original creator. While that sounds like it makes sense now, what happens when the distinction between human and machine disappears? That net will be around us too. Corporations will just use this to empower their copyright rule even further.
Stephen King was largely inspired by Ray Bradbury and H.P. Lovecraft. I doubt he paid them beyond the original price of a couple books.
BTW thanks for the thought provoking conversation. None of my friends care about this stuff 😅
I think we disagree in that you're concerned with something that may happen at some time in the future and how we legislate this event but I see that as a job for future politicians and lawyers.
Right now, I see is a bunch of tech bros with the "move fast and break things" mindset, running around all over the internet and sweeping up data, with no thought to where it came from, if it's legally protected material, sensitive material, or if this might bite them in the ass in the very near future.
theatlantic.com
Hot