Judge tosses publishers' copyright suit against OpenAI
theregister.com
theregister.com
And story writing has literally always been a niche trade, since the dawn of time. A very minuscule fraction of those who do it make it big, and even they start only as hobbyists while having another stable day job.
Most authors today make _almost_ no money. Why would someone who wants to write a book invest months into doing so when they are _guaranteed_ to make no money? Why would anyone go into writing at all in a world where almost all publicized works are created with AI?
Have you ever considered that the entire premise of generative AI "replacing" creatives is misguided and baseless, and that just because it "can", doesn't mean it will? The most it has done to that end is cut off the top lines of corpo stock image and clip art creators. Even where writers use it to create content and reduce their work, it hasn't replaced them altogether.
Having stories? Or having authors?
.. and the mainstream coverage of almost everything in the past half decade. There is no journalism anymore, there is only click and view farming.
Similarly, when I use a local LLM as an extension to my brain by having it process my library using RAG for quick searches and Q&As at a personal level, am I violating it?
Similarly, when my library or university uses it for educational purposes (which literally falls under fair use in most countries where IP law is not a joke)?
And by extension, if a private entity does it with an aim of creating an intelligent public chatbot, as long as it does not monetize by reproducing the author's original content verbatim?
You cannot blurt out such random appeals like asking courts to criminalize training data. Not only is it unfathomably stupid, it is also impossible to enforce. All you'll be doing is cripple your local technological innovation and hand over the edge in something as critical as AI to competing countries in a platter.
When you write this on your new book it it called citation and there are some rules that you should follow so that you are not against the law.
Why are a lot of US companies trying to avoid the law lately ? (Uber, AirBNB, OpenAI)
Evading regulation is profitable. It's even better when your competitors are carefully regulated.
It's part of the business strategy of "regulatory entrepreneurship". Sometimes they can get away with it for a long time, and sometimes they can even change the law.
What did they do that violates what you mention in your comment?
It does not make it automatically right not to do anything these things if you just grab the other people's creation for N times and combining the N results to "form" a new one with some linear algebra procedure, particularly when you also use such synthetic output for a profit. (which is what OpenAI and other companies do).
Facts in a book like a cook book recipe are also not generally copyrightable either. By convention we try not to rip them off but it does not come from copyright.